listedlogmein

EconomicValueoftheAdvertising-SupportedInternetEcosystemJohnDeightonBakerFoundationProfessorofBusinessAdministrationHarvardBusinessSchoolLeoraKornfeldMarlonGerra22017JohnDeighton,LeoraKornfeld,MarlonGerraPublishedbytheauthorsJanuary2017,pursuanttoanagreementwiththeInteractiveAdvertisingBureau,116East27thStreet,7thFloor,NewYork,NewYork10016,whichcommissionedthework.
TheInteractiveAdvertisingBureauwishestothankFacebookandGooglefortheirgeneroussupportofthisresearch.
TheInteractiveAdvertisingBureau(IAB)didnotcontrolordirecttheresearchoritsfindings.
Whiletheauthorsmadeeffortstoassuretheaccuracyofthematerialpublishedinthisstudy,itwaspreparedforgeneralguidanceonmattersofpublicinterestonlyanddoesnotconstituteprofessionaladviceofanykind.
Youshouldnotactupontheinformationcontainedinthispublicationwithoutobtainingyourownindependentprofessionaladvice.
Thisstudymaycontainlinksand/orcitationstothird-partywebsites("ExternalSites").
TheselinksareprovidedsolelyasaconveniencetoyouandnotasanendorsementbytheauthorsorIABofthecontentonsuchExternalSites.
ThecontentofsuchExternalSitesisdevelopedandprovidedbyothers.
YoushouldcontactthesiteadministratororwebmasterforthoseExternalSitesifyouhaveanyconcernsregardingsuchlinksoranycontentlocatedonsuchExternalSites.
WearenotresponsibleforthecontentofanylinkedExternalSitesanddonotmakeanyrepresentationsregardingthecontentorac-curacyofmaterialsonsuchExternalSites.
Youshouldtakeprecautionswhendownloadingfilesfromallwebsitestoprotectyourcomputerfromvirusesandotherdestructiveprograms.
IfyoudecidetoaccesslinkedExternalSites,youdosoatyourownrisk.
TheauthorsandIABmakenorepresentationorwarrantywithrespecttothestudy,whetherexpressorimplied,including,butnotlimitedto,any(i)impliedwarrantyofmerchantabilityand/orfitnessforaparticularpurposeand(ii)anyrepresenta-tionorwarrantyconcerningtheaccuracyorcompletenessoftheinformationcontainedinthispublication.
TheauthorsandIABdisclaimallsuchwarrantiesandrepresentationsand,totheextentpermittedbylaw,theauthorsandIABdonotacceptorassumeanyliability,responsibilityordutyofcareforanyconsequencesofyouoranyoneelseacting,orrefrainingtoact,inrelianceontheinformationcontainedinthispublicationorforanydecisionbasedonit.
Theviews,opinionsandfindingsexpressedhereinarethoseoftheauthorsaloneanddonotnecessarilyrepresenttheviewsoropinionsoftheiremployersoraffiliates.
3SummaryofFindings1.
Introduction1.
1Purposeofthestudy1.
2StructureoftheInternet2.
Methodology2.
1Bottom-upEmployment-BasedApproach2.
2ValueofInternetExportstotheRestoftheEconomy2.
3TimeSpentontheInternet2.
4Top-downEmployment-BasedApproach3.
TheHardInfrastructureoftheInternet3.
1TheForm&FunctionoftheHardInfrastructure3.
2TheHardInfrastructureisevolving3.
3Firmsrespondtothesenewforces3.
3.
1Transmission3.
3.
2Connectivity3.
3.
3Hardware3.
3.
4DataCenters4.
SoftInfrastructure4.
1ServiceProviders4.
1.
1InformationTechnology(IT)Consulting4.
1.
1.
1EnterpriseITConsulting4.
1.
1.
2PublicSector&MilitarySystemsIntegration4.
1.
1.
3OtherInternetITConsulting4.
1.
2Researchers4.
1.
3DomainRegistryServices4.
2SoftwareVendors4.
2.
1AnalyticsSoftware4.
2.
2CustomerRelationshipManagement(CRM)Software4.
2.
3NetworkSecurity4.
2.
4VideoSoftware4.
2.
5AllOtherInternetSoftwareIndex45.
ConsumerServicesSupport5.
1MarketingSupport5.
1.
1AdvertisingAgencies:FullService5.
1.
2AdvertisingAgencies:Digital&CRM5.
1.
3OnlineAdNetworks&Exchanges5.
1.
4Measurement&Analytics5.
2OperationSupport5.
2.
1FinancialSupportServices5.
2.
2E-learningSupport5.
2.
3TravelServicesSupport5.
2.
4HealthcareInformationSolutions5.
2.
5Shipping5.
2.
6E-governmentSupport5.
2.
7WebEnablingServices5.
3GeneralEnterpriseActivity6.
ConsumerServices6.
1ContentSites6.
1.
1News&Information6.
1.
2Multi-genreContent6.
1.
3SpecializedResearch&UserGeneratedContent6.
1.
4OnlineMusicServices6.
1.
5Games6.
1.
6OnlineVideo6.
1.
7E-learning/OnlineEducation6.
2E-commerce6.
2.
1Retailing6.
2.
2TravelServices6.
2.
3FinancialServices:Banking,FinTech,DigitalPayments&Currencies6.
3On-DemandEconomy6.
3.
1Platforms&Services6.
3.
2On-DemandEconomyWorkers56.
4SocialNetworks&Services6.
4.
1SocialMediaSites6.
4.
2OnlineDating6.
5Employment&Recruiting6.
5.
1EmploymentServices6.
5.
2HumanResources6.
6ProductivityTools6.
7Government.
Services7.
IntegratedFirms7.
1ContentMarketersIntegratingVerticallyintoTransmission7.
2TransmissionProvidersIntegratingVerticallyintoContent7.
3IntegratingVerticallyintoCloudTransmissionandDataServices7.
4VerticallyIntegratedFirmConclusions8.
OurConclusionsRegardingEmploymentintheInternetEcosystem8.
1Employment8.
2NewPhenomena9.
DistributionofInternetEmploymentbyGeography:Top-DownApproach9.
1AllocationofInternetEmploymenttoCongressionalDistricts10.
SocietalBenefitsoftheInternetEcosystem10.
1EffectsonNews&Information10.
2EffectsonBusinessModels10.
3MatchmakingforSocietalGood10.
4CommunitiesofInterest:AdHocOrganizing11.
Corroboration11.
1TheInternet'Exports'totheRestoftheEconomy11.
2TimeSpentontheInternet11.
3ComparisonoftheMethods12.
Conclusions12.
1DefiningtheAdvertising-SupportedInternetEcosystem12.
2TheEmploymentandValueofEconomicActivityEnabledbythisEcosystem12.
3TheGeographicDispersionoftheAd-supportedInternetEcosystem12.
4TheCompaniesAssociatedwiththeAd-supportedInternetEcosystem12.
5ContributionofAdvertisingtotheEcosystem12.
6Conclusion6SummaryofFindings1Inourprevioustworeportswedepictedtheinternetasfourhorizontallayers.
Inthisreportweaddafifthelement,agroupofintegratedfirms,forreasonsgiveninChapter7.
Thead-supportedinternethasbeengrowingatabout20%peryearforeightyears:Itsgrowthisaccelerating,and,thoughtwodecadesold,showsthekindofgrowthnormallyassociatedwiththetake-offphaseofanindustry.
Webasethisconclusiononaseriesofstudiesofthisecosystemperformedeveryfouryearssince2008,ofwhichthisisthethird.
Between2008and2012,thenumberofjobsthatrelyontheU.
S.
internetecosystemdoubledfromonetotwomillion.
Between2012and2016,itdoubledagaintofourmillion.
Thereason,asthisreportmakesclear,isthattheU.
S.
economyasawholeisin-creasinglyaninformationeconomy,itsfuelisdata,andtheinternetecosystemisincreas-inglythesystemitreliesonformarket-making.
The2016studyfindsthat4.
1millionpeopleowetheirjobsdirectlytotheadvertis-ing-supportedinternetecosystem,andwhencomparedtothetwoearlierstudieswefindthatcompoundannualgrowthhasacceleratedfrom18.
5percentto19.
6percent,asseeninthefollowingtable.
Foreachpersonwhosejobdependsdirectlyontheinternet,furtherjobs,termedderivedjobs,arecreatedtomeettheirneedsforschooling,entertainment,bankingretail,gov-ernmentservices,andsoon.
Thetablebelowshowsourestimateofdirect,derived,andtotalemployment.
Infrastructure/HardInfrastructure140,000420,000304,393InfrastructureSupport/SoftInfrastructure165,000254,000662,691ConsumerServicesSupport190,000435,0001,068,364ConsumerServices520,000885,0001,619,335IntegratedFirms1442,218Total1,015,0001,999,0004,097,001GrowthinEmployment(%perannumcompound)18.
5%19.
6%Layer2008U.
S.
Employment2012U.
S.
Employment2016U.
S.
Employment7SectorcontributiontoGDPisthesumofnationalincome,corporategrossprofits,andinterest.
Aswasdonein2008and2012,thesectorcontributionwastakentobepro-portionaltosectorincomeusingthenationalaveragewageindexpublishedbytheU.
S.
SocialSecurityAdministration.
Bythismethod,employmentintheadvertising-supportedinternetecosystemcontributedabout$1,121billiontotheU.
S.
GDP,morethandoubleitspreviouscontributionwhenlastcalculatedinthe2012studyat$530billion.
Theindustryaccountedfor6percentoftheU.
S.
GDP,markedlyhigherthanwhenitwasat3.
7percentin2012.
Theeconomicvalueoftheadvertising-supportedinternetecosystemhasgrownatanacceleratingrate,representinga20percentcompoundannualgrowthratefrom2012to2016—faroutpacingtheoverallU.
S.
GDPaverageof4percentoverthesamefouryears.
ThejobsgrowthbenefittedtheentireUnitedStates:Therewasinternet-relatedemploy-mentineverycongressionaldistrictoftheU.
S.
Thedispersionoftheworkforcewasbroad,with116congressionaldistrictsaccountingforhalfoftheemployment,and320accountingfortheotherhalf.
California'sSiliconValleyaccountsforonlyfourpercentofjobsdirectlyattributedtothead-supportedinternetecosystem.
Another10percentisfoundinNewYork'sManhattan,Virginia'sArlingtonCounty,Boston'sRoute128,andWashington'sSeattle/Tacoma.
Butthegreatmajorityofjobs,86percent,liebeyondthecentersofinternetconcentra-tion.
Thisisbecausetheinternetputspeopletoworknotonlyindigitalheadquarters,butwhereverdigitaltechnologyproductsareused–driversmatchedtocustomersacrossthenationonride-hailingplatforms,smallmerchantsineverycityadvertisingonSiliconDirectemploymentduetointernet1,015,0001,999,0004,097,000Directandderivedemployment3,050,0005,100,00010,383,000ContributionofinternettoGDP$300billion$530billion$1,121billionGrowthinGDP(%perannumcompound)15.
5%20.
0%ShareofTotalUSGDP2.
1%3.
7%6.
0%2008Report2012Report2016ReportDirectemploymentduetointernet1,015,0001,999,0004,097,000Derived(indirect)employment2,035,0003,101,0006,286,000Totalemployment3,050,0005,100,00010,383,0002008Report2012Report2016Report8ValleyadnetworksortheclassifiedadsofCraigslistandtransactingonthemerchantplatformsofAmazonandMicrosoft,entrepreneursfundedbyserviceslikeKickstarter,farmersmarketsusingpaymentfacilitatorssuchasSquare,andsocialnetworks,recom-mendationengines,andsearchenginesthathelpsmallsellerswithsmalladbudgetstofindcustomerseventhoughtheylacktheresourcestobuildnationallyrecognizedbrands.
Thecentersofinternetconcentrationarevitalengines,buttheyemployonly14percentofthepeoplewhoearnlivingsfromthead-supportedinternet.
Majorindustrieshaveblossomedinthepast4years:Recreation,retailing,andworkopportunitieshaveallbeenaffected.
Onlinevideoisonesucharea.
Ithasbeenthemostsignificantdriverofnewinternettrafficbetweenthe2012studyandthisreport,asOver-the-Top(OTT)Televisionbe-gantoenterthemainstream.
Thisvideodeliverymechanism—incontrasttotradition-albroadcastandcabletelevisiontransmission—reliesontheopeninternet,suppliedbyahousehold'sinternetserviceprovider,andpermitsunprecedentedcontroloverwhatvideotowatchandwhen.
E-commercealsogrewrapidly,andfromalargerbase.
Onitsownitaccountsfor976,000jobs,almostaquarteroftheentiredirectemploymentbaseofthead-sup-portedinternetecosystem.
Itcontainssomeverylargeandtechnicallysophisticatedenterprises,suchasAmazonandApple,whichalonemakeup26percentofalle-commerceemployment.
Butevenmoresignificantaretheindividuale-commercemerchantswhoselloneBay,Etsy,Craigslist,andAmazon.
Collectivelytheyare29percentofemployment.
Thisthrivingcommunityofworkersillustratestheeconomicopportunitycreatedbytheinternet,notjusttomobilizeindividuals,buttobuildthesoftwarestructurestheyrelyonfortrading,payment,andfulfillment.
On-demandplatformsareathirdnewgrowtharea.
Theseplatformsmatchworkerstoemployersinarangeofindustries.
Part-timeworkersintheso-calledgigeconomy,areinaggregateafull-timeequivalentworkforceof147,000peopleandlikelytogrowmuchlarger.
Usercontentisanumbrellatermforself-publishingoperatingatmuchlargerscalethanatthetimeofthelastreport.
Facebook'sglobalusercommunityhasdoubledto1.
8billionsince2012.
Thereare300millionpeoplepublishingonTwitter,andSnapchat—whichdidn'texistatthetimeofthelaststudy—nowreceivesover10bil-lionviewsperday.
CollectiveknowledgesitessuchasQuora,Reddit,andWikipe-dia,andthevideosof'vloggers'reviewingitemsfromcosmeticstocarstosneakers,haveevolvedfromnoveltiestoregularwebaudiencedestinations.
Thereisanindus-tryofindividualcontentcreators,theiragentsanddistributionnetworksspecializingininfluencermarketing,resultingin,forexample,morethan$1.
5billioninrevenue9and35,000-plusfulltimeequivalentjobsforU.
S.
YouTubecreators.
Music'stransformationfromdownloadingtostreamingisalmostcomplete;Appletalksofdiscontinuingdownloadsaltogether.
Spotify,whichhadatrivialamountofrevenueatthetimeorourlastreport,isnowthedominantproviderofmusicstream-ing,withapproximately100millionusersofwhichcloseto40millionarepayingsubscribers.
Cloudcomputinginthetimesinceourlaststudyhasbeguntodeliversoftwareasaservice,whichallowsittobeconsumedmoreflexiblyandwithoutsubstantialfixedcost.
Itenablesdigitalfirmstotailorinformationtechnologycoststoneeds.
InternetInfrastructuresincethepreviousstudyhasbeguntoevolveitsbackbone-and-branchpatternintoonebuiltarounddatacenters,enablingco-locationofdatageneratorfirmsanddatadistributors.
Inthelastfouryears,particularlyasaresultofthegrowthofdemandofdata-intensiveservicessuchasstreamingvideo,muchoftheinternet'sdatanolongertravelsonthebackbone.
Insteaditmovesontheedgeofthenetwork,fromdatasupplierstodistributionnodesandontohomes.
FirmssuchasVerizonandAT&T,traditionallyhandlingtransmissionanddistribution,haveintegratedintoconsumerserviceandconsumerservicessupport.
Onthehorizonweseethepaceofdata-drivenentrepreneurshipcontinuinginareassuchastheinternetofthings,bots,financialtechnology(FinTech),technology-enabledhealthcare,marketingtechnology(MarTech),andinpublishing,assitesofmediacon-sumptionbecomeuncoupledfromthesitesofproduction.
Allsuggestthatacceleratedgrowthwillcontinuefortheforeseeablefuture.
Changingeverydaylifecreatedatotalof734,000newjobs:Theconsumer-facinglayeroftheinterneteconomy,whichdeliversproductsandservicesdirectlytoconsumers,wasthesegmentwiththelargestgrowthinjobssincethelaststudyin2012.
Asthelayerwherefirmscomefacetofacewithconsumerdemand,itisthelayerwhereeconomicgrowthandconsumervalueoriginate.
Manysmallfirmsandone-personoperationsowetheirjobstotheinternet:Theecosys-tem'slargefirmsaccountedforonly56percentofthe4,097,000jobs.
Mid-sizedandsmallfirms,andasizablenumberofself-employedindividuals,accountedfortherest.
Consumersdidnotpayformanyoftheinternet'sbenefits:Theunder-recognizedsourceofmuchoftheinternet'sconsumervaluewastheinfrastructuresupportingtheconsum-er-facingservices.
Consumerscouldgetthebenefitsoftheinternetatlowercost—andoften,forserviceslikeYelp,Facebook,Twitter,Snapchatandmanyothers,forfree—becauseentrepreneurswerebuildingoutamarket-makinginfrastructuretomakethemprofitablewithoutlargesubscriptionfees.
Forexample,digitaladvertisingagencies,10advertisingexchanges,andmeasurementandanalyticscompanieswhichsupportcon-tentsitesandmarketersseekingtoadvertiseontheinternet,allgrewinresponsetotheincreasingneedtoanalyzeandinterpretthevastamountsofdatageneratedbyconsum-ersoffreeandlow-costservices.
Buildingthisinfrastructurewasanimportantsourceofnewjobs:Oneinfrastructuresectorinparticular,theinformationtechnology(IT)consultingsector,accountedfor12percentofalljobsintheecosystem.
ThissectorcontainsfirmsranginginsizefromglobalgiantslikeAccentureandCognizanttosmallerspecialistfirmsandevenone-personshops.
ITinfrastructureisbecomingaccessibletosmallerfirms:Asinformationtechnologyevolvesfromenterprisesoftwarehostedonafirm'sownedhardwaretocloud-basedservices,opportunitiesemergeforsmallfirmstoapplyITresourcesonceonlyavailabletoglobalgiants.
Astheinternetreducesdependenceonscale,entrepreneurswithgoodideasbutlimitedresourcesbecomeaforcefordisruptiveinnovationintheU.
S.
economy.
Thecontributionofadvertisingtotheecosystemisverylarge:Justwhatdowemeanbytheterm"advertising-supported"Inthepre-internetworld'advertising'referredtopay-mentstomedia.
Ifweapplythisnarrowdefinitionof'advertising'aspaymentstomediaforservicesthatincludesearch,display,classified,mobile,leadgenerationandemail,thenadvertisingcontributed$65billiontotheecosystem.
Butinthepre-internetworldthatdefinition,imposedmainlybyadvertisingagencieswhowerecompensatedinpro-portiontotheirpurchasesofpaidmedia,didnotcoveradvertisingonso-called'owned'mediasuchasdisplaysonthesidesofafirm'strucksandbuildings,nordiditrefertowaystomakesalessuchasdirectmail,catalogretailing,andtelemarketing.
Withthedeclineofcommission-basedcompensationindigitalmarkets,itisunnecessarytodrawlinesbetweenpromotionalactivitiesthatinvolvepaidmediaandthosethatdonot.
Thecommercialactivitiesthatsupporttheinternetareallconcernedwithmarket-making.
Paymentstothird-partymediameasurejustasmallfractionofwhattheinternetdoestomakethemarketsthatcreatetheU.
S.
economy.
Theinternetservesmanypromotionalpurposesbesidesmediaadvertisingnarrowlydefined.
Websitescanserveasbrandedstorefronts,point-of-purchasestimuli,astoolsforconductingresearchonlineforofflinepurchase,andtotransactonlinebasedonresearchoffline.
Websitescanaggregateconsumerreviews.
Consumerscanseeprod-uctspromotedandbuytheminthesamevisitiftheychoose.
Theycandownloaddigitalproductsandconsumethemonline.
Theycansharenewsabouttheirpurchasesandopinionsandreviewproductsandservicesonsocialmedia.
Thereforewhenweusetheterm"advertising-supportedinternet"werefertoallthosein-ternet-basedactivitiesthatmakemarkets.
Theecosystemisineffectadigitalsystemthatmakesmarketsintheonlinesphereandinfluencesmarket-makingintheofflinesphere.
11Itseffects,includingthoseofownedandearnedmedia,arevaluedatits$1,121billioncontributiontoGDP.
Thereisnonewsintheobservationthattheinternethasimpactedcontemporarylifelikenoothertechnicalproductinlivingmemory.
Withthisreportanditstwopredecessorswehavetriedtogomuchfurther.
Wehavequantifieditsimpactonjobgrowthatthreepointsintime,revealedapatternofacceleratingimpactontheeconomy,andhavefoundthemostdynamicofitssectors.
Wehopethatbydeepeningtheunderstandingoftheinternet'sroleintheeconomyanditsstructure,wehaveprovideddataandanaly-sistoinformpolicychoicesastheadvertising-supportedinterneteconomycontinuestoextenditsreachandbecomealargerandmoreintegratedpartoftheU.
S.
economy.
121.
1PurposeofthestudyThisstudywascommissionedbytheInteractiveAdvertisingBureau(IAB)tounderstandthesize,scope,andbenefits,bothsocialandeconomic,oftheecosystemofcom-merceintheUnitedStatesthatowesitsexistencetotheinternet.
ItfollowsthestructureandmethodoftwoearlierstudiesalsocommissionedbyIAB.
Werefertothefirststudyasthe2008studyasitused2008data.
Werefertothesecondstudyasthe2012study.
Itreliedondatafromthesecondhalfof2011andthefirsttwoquartersof2012.
Thisstudy,the2016study,usesdatafromthethirdquarterof2015andthefirstthreequartersof2016.
Incommissioningthesestudies,IABhassoughttotrackovertimethecontributionthatadvertisingontheinternethasmadetotheinternetecosystem.
Itsgoalhasbeentobuildafactbasetodemonstratethevalueoftheindustrytostakeholdersincludingthepublic,advertisers,andpolicymakersinanobjectiveandverifiablemanner.
Thus,IABandtheauthorsshareacommoninterestinobjectivelydeterminingtheextentoftheecosystemthatlivesontheinternet,andthecontributionadvertisingmakestosupportingit.
Likemostinfrastructure,theinternetisfundedinanumberofways:therearefeespaidtoownersandbuildersforaccesstotheinternet,andtherearepurchasesofequipmentandservicestobuildandoperateelementsoftheinternetforprivateadvantage.
Somefundingtakestheformofpaymentsbymarketerstopublishersandothermediafortherighttomarkettothetraffic.
Thesepaymentslessenthecostthateachusermustpaytoreceivethebenefitsoftheinternet,andexpandsthesizeofthesystemthatsocietycanaffordtohave.
Thestudywillestimatethevalueoftheecosystemthatreliesontheinternet,expressedasacontributiontotheUnitedStates'grossdomesticproduct,andestimatethefractionofthatvaluethatisfundedbyadvertisingdefinedbothnarrowlyaspaymentstomediaandbroadlyasmarket-makingservices.
Theassignmentwasto:i.
Definetheadvertising-supportedinternetecosystemii.
Determinetheemploymentandvalueofeconomicactivityenabledbythisecosystemiii.
Determineitsgeographicdispersioniv.
Definethecompaniesassociatedwiththead-supportedinternetecosystemv.
DeterminethecontributionofadvertisingtosupportingtheecosystemWhatistheadvertising-supportedinternetecosystemWhatmarkstheboundarybe-tweenitandtherestoftheeconomyOverthethreestudieswehaveappliedaconsis-Chapter1:Introduction13tentdefinitionofthesystemascommerce-enabling,notthebroadersystemthatcarriesinformationforbothcommercialandnon-commercialpurposes.
Overthesethreestudies,therefore,ourfocushasbeenonactivitiesthatrelyontheinternettopromotemarket-placeexchangeofproducts,services,andinformation.
Whatisthecounterfactual:whatmarketplaceexchangedoesnotbelongintheinternetecosystemObviouslyanalogprocessesareexcluded,suchasthehumanelementsofpersonalselling.
Andacrossthethreestudieswehaveexcludeddyadicdatatransfersperformedonprivatedatanetworksthatwouldhaveoperatedhadtheinternetnotexist-ed.
Butinanincreasinglydata-drivenworld,manymarket-makingprocessesaresupple-mentedbydataflowsandtheydobelonginthesestudies.
Adefiningpropertyoftheinternetisthatitworkswithsoftwarerunningonhigherorlower-layerdatasystemswithoutrestriction,somethingthathasbeentheconsequenceofthesoftwareconventionknownastheInternetProtocol(IP.
)TheIPwasdesignedtobeseparatefrom,andquiteindependentof,thehigher-levelsoftwareapplicationsthatrunonit,creatinganopenplatformthatcancarrydataofanykind,whetherwrittentobereadbyemail,filesharing,videostreaming,orwebsiteassemblysoftware.
(TheWorldWideWeb,forexample,whichisoftenconfusedwiththeinternet,isoneofthesehigher-layerapplications,runningontheHTTPprotocol.
)Theinternetwasalsodesignedtobeprovisionedovermanylower-layerphysicalnetworkssuchascoaxialcable,Eth-ernets,telephonenetworks,cablenetworks,andwirelessnetworks,withouttheneedtotailorthem.
TheIP,insummary,hasbeensucharemarkableeconomicforcebecauseitcarriesmanykindsofdataonmanyphysicalarchitectures,almostwithoutrestriction.
WhatisrestrictiveabouttheInternetProtocolis—asRobertCannon2hasargued—itsuniqueaddressspace.
Thisaddressspaceisdefined,andwithittheinternetcanbedefined,astheuniquesetofIPnumberaddressesassignedbytheInternetAssignedNumberAuthority(IANA),adepartmentoftheInternetCorporationforAssignedNamesandNumbers(ICANN),anon-profitcorporationbasedinCalifornia.
AsFigure1-1illustrates,theIPisjustonelayerinastackofprotocolsadoptedtotrans-mitdatafromvariousphysicalnetworkstovarioussoftwareapplications,butitisaparticularlyimportantlayer.
Thestackhasanhourglassshapewithaverydistinctneck,sothatmanykindsofdatapassthroughtheIPlayerofthestack.
TheIPenablespacketswitching,andisdistinctfromandindependentofvoicenetworksthatrelyoncircuitswitching3.
ThustheIPlinksnetworkscarryingdigitalinformation,andempiricallyiseasilydistinguishedfromnetworkscarryinganaloginformation,includingvoicetransmissions.
2Cannon,Robert(2014).
"WilltheRealInternetPleaseStandUp:AnAttorney'sQuesttoDefinetheInternet"(March2004).
TelecommunicationsPolicyResearchConference2002.
AvailableatSSRN:http://ssrn.
com/abstract=516603orhttp://dx.
doi.
org/10.
2139/ssrn.
516603.
3SeeforexampleKahnandCerff,"WhatistheInternet"http://www.
cnri.
reston.
va.
us/what_is_internet.
html14Figure1-1:InternetArchitectureWeusetheterminternetecosystemtodescribetheobjectofthisstudy:anaggregationofbusinessesthatdependsonandco-evolveswiththeinternetinfrastructure.
Thetermdoesnothavealongpedigreeineconomics,butitisusefulhere.
Itderivesbyanalogyfromaunitofanalysisinbiology,whereitreferstoaninterdependentsystemoflivingorganisms,fromplantsandanimalstomicro-organisms,takentogetherwiththeinertelementsoftheirenvironmentsuchaswaterandsoil.
Althoughthetermhasbeenusedinbusinessandeconomicssinceatleastthe1970s4,ithasbecomepopularinmorerecentyearsintheinformationtechnologyindustry,wheretheinterdependenceofbusi-nessesrelyingonacommonsetoftechnologiesisacentralfact.
Moore(1996)wroteofabusinessecosystemasa".
.
.
community[offirmsthat]co-evolvetheircapabilitiesandroles,andtendtoalignthemselveswiththedirectionssetbyoneormorecentralcompa-nies.
Thosecompaniesholdingleadershiprolesmaychangeovertime,butthefunctionofecosystemleaderisvaluedbythecommunitybecauseitenablesmemberstomovetowardsharedvisionstoaligntheirinvestments,andtofindmutuallysupportiveroles.
5"ForwhatapplicationisthedatausedWebsiteseMailFilesharingStreamingHTTPWhatprotocolsareusedHowisthedatacarriedPOPSMTPP2PRTPCoaxialcableTwistedpaircopperwireOpticalfiberWirelessTCPUDPIPPPPEthernet802.
11DOCSIS4Boulding,K.
(1978).
Ecodynamics:ANewTheoryofSocietalEvolution.
BeverlyHills,CA:Sage.
5Moore,JamesF.
(1996).
TheDeathofCompetition:Leadership&StrategyintheAgeofBusinessEcosystems.
NewYork:HarperBusiness.
15ThestructureofabusinessororganizationalecosystemhasbeeninvestigatedbyMars,BronsteinandLusch.
6Theyfindthatplayersinteractinpatternsthataresometimesmutuallyadvantageous,sometimescompetitive.
Theyargue(asdoIansitiandLevien7)thatsomeplayersactaskeystones,inthesensethattheyoccupyhubsintheexchangenetworkwhosehealthassuresthehealthofthewholesystem.
Theecosystem,theyfind,isanestedstructureinwhichfunctionsandprioritiesoftenoverlap.
Theseredundanciescreateresilience,andconverselytheirabsenceputsthesystematriskofcollapse.
Theyarguethatorganizationalecosystems,unlikemostbiologicalecosystems,haveforesightandcananticipateconditionsthatmightleadtosystemcollapse.
Themorecomplexthesystem,andthemoreturbulentitsevolutionarypath,however,themoredifficultitistounderstandtheinterdependenciesthatputthesystematrisk.
Finally,theyemphasizetheemergentnatureofbusinessecosystems.
Whilestrategyanddeliberatedesignareneverirrelevant,theyarenotthedeterminants.
Anecosystemcannotbedesignedintoexis-tence,althoughalegalregime,aregulatoryregime,education,andaccesstocapitalcanbedesignedtoenableitscontinuity.
Onequestionthestudyseekstoexploreistheextentoftheecosystem'srelianceonadvertisingtosupportit.
Advertisingcanbereadnarrowlyaspaymentsbyadvertiserstopublishers,followingtheprecedentestablishedinthepre-internetworld.
Inthatworld'advertising'didnotcoveradvertisingonso-called'owned'mediasuchasdisplaysonthesidesofafirm'strucksandbuildings,nordiditcoverdirectmail,catalogretailing,ortelemarketing.
However,inthedigitaleconomy,thisdistinctionunderplaysoneoftheimportanteconomicconsequencesoftheinternet.
Themarketingeffectsoftheinternetecosystem,particularlythoseofownedandearnedmedia,areverysubstantial.
Pay-mentstopublishersdonotmeasureallthattheinternetdoestomakethemarketsthatcreatetheeconomy.
Theinternet,insum,servesmanycommercialpurposesbesidesadvertisinginthenar-rowsenseoftheword.
Websitescanserveasstorefronts,point-of-purchasestimuli,astoolsforconductingresearchonlineforofflinepurchase,andtotransactonlinebasedonresearchoffline.
Websitescanaggregateconsumerreviews.
Consumerscanseeproductspromotedandbuytheminasinglevisit.
Theycandownloaddigitalproductsandconsumethemonline.
Theycansharenewsabouttheirpurchasesandopinionsandreviewproductsandservicesonsocialmedia.
Reflectingbacktothepolicygoalofthestudy,weexploretheinternet'smanyadvertis-ingbenefitsbesidesthosethatcomefromnarrowly-construedmediaspending.
Inaddi-tion,weexploretheinternet'snon-businessbenefits.
6Mars,M.
,Bronstein,J.
andLusch,R.
(forthcoming).
Thevalueofametaphor:Organizationsandecosystems.
OrganizationalDynamics.
7Iansiti,MarcoandLevien,Roy(2004).
TheKeystoneAdvantage:WhatthenewDynamicsofBusinessEcosystemsMeanforStrategy,InnovationandSustainability.
HarvardBusinessSchoolPress.
161.
2StructureoftheInternetOurearlierstudiesdescribetheoriginsoftheinternet,andweshallnotrepeatthatmaterialhere.
Greenstein(2015)hasanexcellentandmuchricheraccount.
8Themod-elusedinthisstudyfollowsthatusedinthe2012report:atreestructure,whosehardinfrastructureisanalogoustoroots,asoftinfrastructuretrunk,consumerservicessupportmakingthebranches,andconsumerservices,likeleaves,correspondingtotheservicesthatconsumersbuyorreceiveinexchangeforadvertisingservices.
Thetreemetaphorconveysthelogicoftherelativestabilityoftherootsandtrunkrelativetotherichnessofthefoliage.
Since2012,however,wehaveintroducedanintegratedcategorytocon-tainfirmswhosescoperunsfromrootstoleaves,aswewilldiscussinChapter7.
Figure1-2:StructureoftheInternet2016ConsumerServicesConsumerSupportServicesIntegratedFirmsSoftInfrastructureHardInfrastructureContenteCommerceOn-demandEconomySocialNetworksBusinessServicesGovernmentServicesMarketingSupportOperationSupportSoftwareServicesTransmissionConnectivityHardwareDataCenters8Greenstein,Shane(2015).
HowtheInternetBecameCommercial.
UniversityPress:Princeton,NJ.
17Anation'sgrossdomesticproduct(GDP)istheaggregateofincomesreceivedbyresidents,bothindividualandcorporate,asdirectpaymentforcurrentservicestoproduction,plusreturnoncapital.
Thisisequivalenttothesumofthevaluesaddedateachstageofproductionbytheindustriesandproductiveenterpriseslocatedin,andmakingup,thecountry'sdomesticeconomy.
ThenationalGDPcanbedecomposedintosectorGDPs,which,inthesamespirit,measuretheeconomicactivityofeachsector.
ThisapproachestimatestheinternetsectorGDP.
Ourobjectiveistoprovidedatathat,togetherwithreasonableassumptions,canbeusedtoestimateU.
S.
economicactivityattributabletotheadvertising-supportedinternetecosystem.
Thisreport,asthe2008and2012studiesdid,usesanemployment-basedmethodology,andtwomethodstotesttheplausibilityoftheanswer:1.
Bottom-upEmployment-BasedApproach:Fromavarietyofsourcesweidentifylargefirmsineachlayeroftheinternet,andusearangeofpublicandprivatesourcestoestimateeachfirm'srevenueandemploymentaswellasestimatesofaggregatesofsmallfirmsandself-employedpeople.
Thismethodisprecisewithrespecttothefirmsstudied,butapproximatewithrespecttorelevantemployment.
2.
WhattheInternetExportstotheRestoftheEconomy:Wecomputepaymentsbyfirmsforinternetservices,viewingtheinternetasifitwereanislandexportingtotherestoftheeconomy.
3.
TimeSpentontheInternet:Wevaluethetimethatusersspendontheinternetatitsmarginalvalue.
Inaddition,weuseatop-downmethodtodecidehowtoallocatetheemploymenttogeographies.
4.
Top-DownEmployment-BasedApproach:FromtheU.
S.
CensusBureau'sdatabas-esweidentifycategoriesofestablishmentwithsignificantinternet-relatedemploy-ment.
AstheU.
S.
CensusBureauprotectstheconfidentialityofitsrespondentfirms,thismethodisprecisewithrespecttoemployment,butapproximatewithrespecttofirmcontributionstotheinternet.
2.
1Bottom-UpEmployment-BasedApproachThisapproachuses,asastartingpoint,theNorthAmericanIndustrialClassificationSystem(NAICS)codesoffirmsstudiedinthe2012report.
ForeachNAICScodeweChapter2:Methodology18identifythelargestemployers,theirtotalrevenueandemployment,andtheproportionofrevenueandemploymentthatweattributetotheinternet-relatedactivitiesofthefirmintheUnitedStates.
Thefirm-by-firmassumptionsthatunderlietheseattributionscanbeobtainedbyrequestingaspreadsheetfromtheIAB.
Inthisreportwereviewtheassump-tionsbehindthelargerandmostmaterialoftheestimates.
Inaddition,wemakeallowanceforsmallerfirmsbycomparingourfirm-by-firmenumer-ationstothenumberofemployeescountedinU.
S.
BureauofCensusreportsandtradeassociationindices,andwherenecessarywecreate'allother'categories.
Forindustrialactivitythatistoonewtohavebeencountedinthemostrecentbusinesscensus,suchasmobilephoneapplicationdevelopment,wemakespecialadjustmentsdescribedinthebodyofthereport.
Weallowforsectorscomprisingfirmsthatwereindividuallysmallbutlargeinaggre-gate,namelylocalandstatee-governmentservices,smallonlineretailers,andpeopleworkinginlargeandmid-sizedgeneralenterprisesperformingworkontheinternetbutnototherwisecountedintheinternetecosystem.
Finally,wecountself-employedworkerssuchassellersonEtsyindividualsellersoneBay,on-demandeconomyworkers,andfreelanceindividualsdoingcoding,contentcreation,andotherservicesforwebsites.
Foreachpersondirectlyemployedinaparticularsectoroftheinternetecosystem,otherpeopleworkinsectorsthatsupplythesectororthatbenefitfromretailandservicesectorspendingbytheseworkers.
Thefocalsectoralsohelpstosupporttaxation-dependentareasoftheeconomy,suchasgovernmentandpublicsectorworkerswhoareemployedinfederal,state,andmunicipalservices,education,andthemilitary.
Thus,thisindirectemployment,computedbyapplyingemploymentmultiplierstothesector'semployment,arisesfromsuppliereffects,re-spendingeffects,andgovernmentemploymenteffects.
TheU.
S.
BureauofLaborStatisticspublishesstatisticsonindustryemploymentrequirements,whichenablecalculationofthelaborinputsintoasector.
Sectorsdifferinthesizeoftheirmultipliers.
Bivens8computesindirectemploymentthatrangesfrom372indirectjobsforevery100jobsindurablesmanufacturingto163indirectjobsforevery100jobsinbusinessservices.
Theseestimatesareinclusiveofcapitalserviceusage.
Wetaketheappropriateratiofortheinternetsectoras154indirectjobsforevery100jobsdirectlycreated,anchoringonthebusinessservicesratioandadjustingforconsumerservicesandinfrastructurefirms.
Wethenapplyafullyburdenedlaborcost,comprisingwagesandsalaries,thecostofbenefits,on-boarding,managementoverhead,vacationtime,andfacilitiescosts,totheseemployees.
8Bivens,Josh(2003)UpdatedEmploymentMultipliersfortheU.
S.
Economy,EPIWorkingPaper268.
EconomicPolicyInstitute.
192.
2ValueofInternetExportstotheRestoftheEconomyThedirecteconomicvalueoftheservicesthattheinternetprovidestotherestoftheU.
S.
economyistherevenuepaidfortheservices'exported'beyondthebordersoftheinternet'seconomytotherestoftheU.
S.
economy,netofwhatis'imported'.
Themajorcategoriesofexportcompriseadvertisingservices,retailtransactions(netofcostofgoods)conductedontheinternet,anddirectpaymentstointernetserviceproviders.
Inaddition,theinternetgeneratesanindirecteconomicvalueofactivitythattakesplaceelsewhereintheeconomyduetotheinternetsector.
Thesamemultiplierisusedaswasusedforemployment.
2.
3TimeSpentontheInternetThethirdmethodisbasedonthetimethatpeople'give'totheinternet.
Werelyonanumberofstudiesofinternetuse,someofwhichweresurveysofrecalledbehaviorandothersthatwerebasedonobservationofactualbehavior.
WeestimatethevalueofanhourspentatworkforarepresentativeU.
S.
workerat$19.
07perhour,derivedfromtheaveragewageofnon-management,non-agriculturalworkersindatapublishedbytheBureauofLaborStatistics.
Thereisnomarketpriceforanhourspentinrecreationorleisure,althoughthereisanopportunitycost.
Ifworktimeisdiscretionary,thenithasbeenargued(Bockstaeletal.
,198710)thatthewageratemeasurestheopportunitycostofleisuretime.
Ifnot,andinparticularforpeopleinschoolorunder-employed,thewagerateover-estimatesthevalueofaleisurehour.
Asanapproximation,weuse10percentofthewagerateforleisuretime.
2.
4Top-downEmployment-BasedApproachThegoalofthismethodologyistomapthelocationofemploymentinindustriesthatarepartoftheinternetecosystem,tosupplyabroaderdistributionofemploymentthanisavailablefromthelocationofheadoffices.
WeusetheU.
S.
CensusBureau'sCountyBusinessPatterns(CBP)dataset,whichgivesthenumberofemployeespercounty,bothcoreandsupport,foreachestablishmentineachoftheapproximately700five-digitcodesoftheNorthAmericanIndustryClassificationSystem(NAICS).
FromadetailedreviewoftheindustrydefinitionsofeachNAICScode,weidentify15ofthe700thatarelikelytohavemeaningfulamountsofinternet-dependentemployment.
Threeareentirelyinternet-dependent,and12arepartiallydependent.
Forthelatterwecalculatean'internet-intensity'ratioforitsNAICScode.
Fromourbottom-updata,whichincludeseachfirm'sNAICScode,wecomputethecode'sinternetrevenueasafractionoftotalU.
S.
revenue.
Forcodeswhereourbottom-upanalysisdoesnotyieldarepresen-tativesampleofcompanies,weutilizeProductLineReceiptsfromthe2012Economic10Bockstael,N.
,Strand,I.
andHanemann,W.
(1987),"TimeandtheRecreationalDemandModel,"AmericanJournalofAgriculturalEconomics.
69(2)293-302.
20Censusasanindicatoroftheproportionofrevenuesourcedfrominternetwork.
Wedidnotrelyonthismethodtocorroboratethebottom-upmethodologybecausethetwomethodsarenotentirelyindependent,andbecause,asdiscussedbelow,whileitisagoodmeasureofthegeographicdistributionofemployment,itisnotareliablemeasureoftotalinternetemployment.
21Internettrafficworldwidehasgrownrapidlyinrecentyearsandisexpectedtocontinuetogrow.
ItisreportedbyCisco'sVisualNetworkingIndexthatthatthevolumeofmo-biledatatrafficin2015hadgrownby15timessince2010.
ThesamereportforecaststhatglobalIPtrafficwillgrowfrom72.
5exabytespermonthin2015to194.
4exa-bytesin2020,anannualcompoundgrowthrateof22percent.
InthischapterweidentifythefirmsandemploymentthatmakeupwhatwecalltheHardInfrastructurelayeroftheinternet.
WeusethistermtodistinguishitfromSoftInfrastruc-ture,wheretheproportionofsoftwareandservicetohardwareismuchgreater.
3.
1TheFormandFunctionoftheHardInfrastructureIntheHardInfrastructurelayeroftheinternetweidentifyfourfunctionsrequiredtooperatetheinternet:manufactureofhardware,longrangetransmissionofdata,shorterrangeconnectivitybetweendatageneratorsortransmittersanddatastoresorconsum-ers,andthestorageandroutingofdataindatacenters.
Thesefourfunctionsgenerateemploymentinfirmsthatspecializeinoneofthefunctions,infirmsthatconcentrateononebutperformothers,and,increasingly,infirmsthatperforminfrastructurefunctionsbutareintegratedacrosstheentireinternetstack.
InthislattercasewehaveanalyzedthefirmsinChapter7,IntegratedFirms.
Whentheinternetwascommercializedinthelate1990s,anumberoffirmsinthetelecommunicationsindustrypursuedthetransmissionopportunity.
TheyinvestedheavilytolayfiberopticcableacrosstheUnitedStates,Canada,Europe,Asia,andtheoceansthatseparatedthem.
Transmissioninfrastructurewasradicallyoversuppliedforthefirstdecadeofthiscentury,includingatthetimeofthefirstofourstudiesoftheecosystemin2008.
Voiceanddial-upmodemtransmissionofdataplacedlittledemandontheinfrastructure.
Bandwidth-intensiveservicessuchasvideoandmusicstreaming,softwareasaservice,mobilebroadbandtransmissionamongcellphonetowers,andbigdataanalytics,begantoemergeduringthesecondphase,whichbeganinseconddecadeofthecentury,andbegantodeliverslightlyimprovedeconomicreturnstothefirmsthathadbuiltthefiberinfrastructure.
However,thepatternweshallreportisoneinwhichinfrastructurefirmsstruggletocommandtheprofitabilityofotherinternetlayers.
Infrastructureemploymentisdecliningnotbecausetheinfrastructureisshrinking(itisnot,witnesstheremarkableincreasesintrafficjustreported),butbecausefirmsthatwereclassifiedasinfrastructurejustfourChapter3:TheHardInfrastructureoftheInternet22yearsagohavesochangedtheirbusinessmodelsthatwehavehadtoclassifythemelsewhere.
Somehaveintegratedintomoreprofitablesuperstructurebusinesses,hopingtocapital-izeonproprietaryclaimsontheinternettrafficcarriers.
Wetreattheminanewchapter,IntegratedFirms.
Othershaveexploitedanewpatterntotheinternet'sinfrastructure,onethatdependsonsoftwareandservicesmorethanhardware.
3.
2TheHardInfrastructureisEvolvingThisnewinfrastructureistermedthe'cloud'.
Itcontrastswiththetrunkandbranchformofthefirstphase,whichcomprisedatrunkmadeupofanintertwinedandcross-con-nectedsetoflong-haulfibertransmissionlinesknownasthebackbone,andbranchestodistributedatatotheultimateusers.
Thedistinctionbetweentransmissionanddistributionfunctionsisacommonfeatureoftheinfrastructuresofutilitiessuchaselectricalpower,telephony,andwater,explainedbythefactthattransmissiontrunks,becausetheycarryoverlongdistances,areexpensivetobuildbutinexpensivetooperate,withfewcustom-erstobill,whilethedistributionbranchesaredensestructureswithmanycustomerstoserveandbill.
Inour2008and2012reports,mirroringthisfirstphase,weclassifiedemployersintotransmissionanddistributiongroups.
Transmissionfirmswerethen(andtoanextentstillare)thosethattransmitteddataoverlong-haul(inter-city)andshort-haul(within-city)distances.
Transmissionsystemswerecommonlysharedamongfirms,andthelargestfirmsenteredintotransit-freepeeringarrangementswithoneanother.
Distri-butionfirmswere(andtoanextentstillare)firmsorganizedbytheconnectiontechnol-ogy—cable,phone,andwireless.
Theymoveddatafromthetransmissionsystemstohouseholdandbusinesssubscribersandwerecommonlynotshared.
Indeed,asinglesubscribercouldhaverelationshipswithseveraldedicateddistributors,suchasacablesubscription,awirelesssubscriptiontoamobilephone,andoccasionallyasubscriptiontolinkanautomobiletotheinternet.
Inthe2008and2012reports,thethreefunctionsofhardware,transmission,anddistribution,mappedrelativelywellontothreekindsofemployer:thebigtelecommunicationsfirms(telcos)handlingtransmission,cableopera-tors,smallertelcosandwirelessoperatorshandlingdistribution,andthemanufacturersmakinghardwareandequipment.
Atthetimeofthisstudywearebeginningtoseethebackbone-and-branchpatternofthelasttwodecadesgivewaytothemorefluidpatternofthecloud.
Threefactorsareatworkinthisevolution,disruptingtheseparationoftransmissionandconnectivityandindeedbeginningtodisrupttheseparationofinfrastructurefromconsumerservices.
First,asconsumershavesoughtmorethanonewaytoconnecttotheinternet,itwasattractiveforconnectivityfirmstoorganizebycustomerandgeography,notbytechnolo-gy.
Cableoperatorsofferedvoice-overIPphoneservice,andwirelessandcablemodem23internetconnectivity.
Telcosenteredthewirelessandopticalfibermarketsandlargelydisplacedinternetconnectivityovercopperwire.
Theopeningupofmillimeterwavespectrumpresentstheprospectofwirelesssystemsthatcancarrylargevolumesofdata.
Googleisbackingawayfromitsrolloutoffibertransmission,anticipatingawirelessoptionwithmuchlowerinstallationcosts.
Severalofthetransmissionfirmsnowhavesignificantconnectivitybusinessesandasthetwofunctionsintegrate,thebackboneandbranchpatternbecomeslessdiscernable.
Second,consumersbegantousedatainvolumestoogreatforthebackbonetohandlewithtransit-freepeering.
TheysoughtconnectiontohighbandwidthcontentproducerssuchasNetflix,YouTube,andFacebook,particularlyatthetimesofdaythatusedtobecalledprimetime,whenmanyhomesviewedvideo.
Distributionfirmssoughtwaystomovethisdatadirectlytotheirhouseholdsubscriberstoenhancetheiruserexperienceandstrengthentheloyaltyofsubscriberstothem.
Thesedistributorswouldco-locateNet-flix'sdatatransmissionservices,forexample,intheirowndatacentersatnocosttoNet-flix,andmovingtrafficfromlongdistancetransmissionservicestodistributionservices.
Thusfarfrompayingtransmissionfirmstocarrytheirdata,suppliersofcontentvaluedbyhouseholdsfoundthatdistributorswouldcarrydataflowsatnocostintheinterestsofwinningtheloyaltyoftheircustomers.
Theseso-called'edgeproviders'becamepartofthefourthfunctionmentionedatthestartofthissection,datacenters.
Datacentershavelongbeenanelementoftheinternet'sinfrastructure,butinthepost-backboneeraoftheinternettheirfunctionismorefundamentaltotheinternet'sin-frastructure.
They,whetherintheformofgiantserverfarmsortinyco-locationfacilities,arethephysicalexpressionoftheideaofthecloud.
Datacenters,enablingco-locationofdatageneratorfirmsanddatadistributors,givetheinternetinfrastructureitsnewshape.
Inthelastfiveyearsorso,particularlyasaresultofthegrowthofdemandofdata-intensiveservicessuchasstreamingvideo,muchoftheinternet'sdataisnolongertransmittedonthebackbone.
Insteadthisdatamovesontheedgeofthenetwork,fromdatasupplierstodistributionnodesandontohomes.
Inthisreportweusethetermdatacenterstorefertothelocationofjobsinfirmsthatsupplytheseedgeflows.
Thelargestdataprocessors,Amazon,Google,andFacebookowntheirownserverfarms,andlargedatacenterfirmsincludingAkamai,Equinix,Rack-space,andtheEuropeanfirmInterxion,locateatpointsofhighretransmissiondemand.
Athirdgroupfillsregionsoflessdensedemand,includingEdgeConneXandCloudflare.
3.
3FirmsRespondtoTheseNewForcesInjusttheyearsbetweenourlaststudyandthisone,firmsthatweretraditionallyre-sponsiblefortransmissionanddistributionhaverespondedtothecommodificationofthisworkbyintegratingintoconsumerserviceandconsumerservicessupport.
Verizon,24whoserootsareinthephonecompanyBellAtlantic,acquiredtheconsumer-facingfirmsAOLandYahoo.
CharterCommunicationsandCoxfollowedsimilarpaths.
Aswewrite,thetelcoAT&T,havingacquiredsatellitedistributorDirecTV,isattemptingtowinregula-toryapprovaltoacquirethecontentproviderTime-Warner.
Andfirmsbornatthecon-sumer-facingendofthetraditionalinternetstructuresuchasAmazonandGoogleareintegratingbackintotheinfrastructurebusiness.
Asnotedearlier,wedevoteachaptertothesefullyandpartiallyintegratedinternetfirms.
WewillsummarizeemploymentintheInfrastructureLayerunderthefourfunctionheadings,commentingwhereappropriateonfirmsthatcrosstheboundariesbetweenfunctions.
3.
3.
1TransmissionTwoofthesixfirmsclassifiedasTransmissionfirmsin2012arenowanalyzedintheIntegrationchapterbecause,althoughtheystillcontributetotransmission,theyhavesoughtrevenueopportunitieselsewhereintheinternetecosystem.
TheyareAT&TwiththeacquisitionofDirecTV,anditsbidforTime-Warner,andVerizonwiththeacquisitionofAOLandthepotentialacquisitionofYahoo.
Fourothers,Sprint,Level3,InteliquentandCenturyLink,remainsubstantiallyintransmission.
Threefirmswithlimitedpreviousinvestmentintransmission,Alphabet(Google),Face-bookandMicrosoft,arenowimportantelementsofthetransmissioninfrastructure,alsoanalyzedintheIntegrationchapter.
Theirunderseafiber-opticcablescreateprivate11dedicatedcapacitythatlinkstheUnitedStateswithAsiaandEurope.
OurestimateforTransmissionProvidersis$25.
7billionand43,269jobs.
Table3.
3.
1:TransmissionProviders11Finley,K.
(2016).
"FacebookandGoogleWillStretchInternetCablefromLAtoHongKong.
"Wired,Oct12,2016.
Sprint$11,10910,356CenturyLink$9,76823,465Level3Communications$4,2408,466CogentCommunications$404836Inteliquent$2331662016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees25Notethatthissegmentissmallerthaninthe2012report.
Theconclusionisnotthattransmissionemploymenthasdeclined,butthatfirmsthatsupplytransmissionmustseekprofitableoperationsinadjacentmarketsandso,inthefouryearssinceourlaststudyhavecometobeclassifiedindifferentways.
3.
3.
2ConnectivityAsdatastoragehastransitionedfromlocalstorageonafirm'sownserverstodistributedstorageonawidelydispersed,oftenglobal,networkofdatacenters,connectivity'sroleintheinternet'sinfrastructurehasgrown.
Dataoncestoredondiskandbackedupontotapeisnowoftenbackedup,managedandstoredonavarietyofsystemsthatrangefromdisktotapetopublicandprivatecloudstorageservices.
Theevolutionhasbeendrivenbygrowthinthequantityofdatausedinbusinesses,inthedecliningcostofdiskstorage,andthedemandforredundancyindatastorage.
Storageincreasinglytaxestheinternet.
Connectivityfirmsprovidesometransmission,andtransmissionfirmsofferconnectivity,sothereisadegreeofarbitrarinessinthedistinctionwearemakingbetweenthetwo,butourgeneralruleisthatfirmslistedhereretailconnectivitytomuchsmallerclientsthandothetransmissionfirms,particularlyhouseholds,andoftensellitasacomple-menttomoreprofitableservicesdeliveredontheirpipes,suchascabletelevisionandwirelesstelephony,ortointernetservicessuchasemailandspamprotection.
Whereonceconnectivitywasenvisionedasanopportunitytobecomeaportalwithproprietarycontent,itiscurrentlyviewedasanearcommodity.
OurestimateforConnectivityProvidersis$27.
4billioninrevenueand48,986jobs.
26Table3.
3.
2:ConnectivityProvidersCableProvidersCharterCommunications$3,5834,958Altice(formerlyCablevisionSystemsCorp.
)$1,2902,798Mediacom$270753Shentel$4494InternetAccessWindstreamHoldings$3,4187,307TelephoneandDataSystems$1,9931,510ZayoGroup$1,5081,606EchoStar$1,3471,885HarrisCapRockCommunications$8593,936CincinnatiBell$8062,243BirchCommunications$6801,360ConsolidatedCommunications$4711,083Gogo$5011,073GCI(GeneralCommunications,Inc.
)$400969DISHNetwork$439545HawaiianTelcom$149149NovatelWireless$140600FTC$35100HTC(HorryTelephoneCooperative)$6464MobileInternetAccessProvidersT-MobileU.
S.
$6,50113,095U.
S.
Cellular$2,6743,200TracFoneWireless$149145IridiumCommunications$3030iPass$24832016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees27In2012,wecountedConnectivityasresponsiblefor91,000employees.
In2016wefindonly49,000.
However,another45,000workers,morethantheentiredifferencebetween2012'semploymentand2016's,isaccountedforbythefactthatwemovedComcast,CoxCommunications,andTimeWarner,outofHardInfrastructureandintootherlayers,becauseacquisitionsandrevenuegrowthinnon-connectivitysegmentsoftheirbusinesseschangedourassessmentoftheircorebusinesses.
Insum,inthissegmentasintransmission,theworkershavenotbeenlost,buttheiremployershavepursuedbroaderstrategiesandwehavehadtoclassifythemdifferently.
3.
3.
3HardwareInternet-relatedhardwarecomprisescomputers,servers,storagedevices,routers,switch-es,desktopandlaptoppersonalcomputers,wirelessaccessdevices,fiberopticcable,andbroadbandwirelessequipment,amongothercomponents.
Overtheyearstherehasbeenconsolidationamonghardwaresupplierssothattodayhardwareisoneofthemostconcentratedoftheinternetsegments.
Therehasbeenashiftinthelastdecadefrommanufactureofcomputingequipmenttodatahandling,switching,andstorage,withthegrowthindatacenters,thecloud,andtheshiftfromalineartoanetworkedstructure.
Forexample,the2015annualreportofCrayInc.
notesthatamongitscustomers,"Data…andcapacityneedsaregrowingmuchfasterthancomputationalneeds.
"Atthesametimepersonalaccesstotheinternethasmigratedrapidlyfrommachinesconfiguredforcomputingtomobiledevicesconfig-uredforcommunication,transforming,forexample,Applefromalowsharemanufac-turerofcomputerstothegloballeaderbyrevenueinthemobiledevicesmarket,andcomingattheexpenseofdesktopmanufacturers.
Demandformobiledeviceshasbeendriveninpartbyconsumerswhousethemforvid-eoviewingorretransmissionofvideostreamstowidescreenmonitors,underthegenerallabelofover-the-top(ofthecableprovider)services.
ThisdemandisfedbyvideosupplyfromnewmediacompaniessuchasNetflix,Hulu,Google(YouTube),Amazon(AmazonInstantVideo)andApple(iTunes)whoseemploymentiscapturedinotherchaptersofthisreport.
U.
S.
-basedemploymentinthissectorismoreengagedinproductdesign,software,marketing,salesandservicethanmanufacturing.
Forexample,theU.
S.
haslostemploy-menttoAsianfactories(e.
g.
halfofthe23,000employeesofCommScopearelocatedoutsidetheU.
S.
).
Ourestimatesoftheproportionofemploymentduetotheinternetaresensitivetotherevenuesoftheirlinesofbusiness,andinsomecaseswereceivedguid-ancefrompeopleinthecompaniesorindustries.
Ourestimatesareconservative.
OurestimateforHardwareis$159.
3billionand196,905jobs.
28Table3.
3.
3:HardwareHardwareemploymentrosefrom148,000in2012to196,000thisyear.
ThisgrowthisdespiteHewlettPackard(23,000employees)andIBM(36,000employees)movingtothenextchapterbecausetheynowdomoreworkinconsultingthaninmanufacturing.
Thelastfouryearsoftheevolutionoftheinternetecosystemhavefavoredmobileequip-mentmanufacturerslikeAppleandQualcommandthoselikeCiscothathavebenefitedApple$74,47839,600CiscoSystems$26,69036,222Qualcomm$18,20223,760Dell$18,01348,034NetApp$3,0676,653Ericsson$2,7684,982JuniperNetworks$2,2794,529CommScope$1,91010,350F5Networks$1,4403,134Seagate$1,3501,880Supermicro$1,161420Avaya$1,0402,984Brocade$5661,160NimbleStorage$3001,000AcerAmerica$256120ExtremeNetworks$239583UbiquitiNetworks$200180RuckusWireless$179480Harmonic$160541PureStorage$134650DigiInternational$106444Cray$5020CSPEmbeddedComputer$45862016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees29fromgrowthinserversales,notmainframecomputers.
Manufacturingemploymenthasheldupwelloverall.
3.
3.
4DataCentersAsdiscussedearlierinthischapter,theinfrastructureoftheinternetisgraduallymigrat-ingfromalineardataflowpatterntoanetworkedpattern,wheredatatravelsasmuchamongfirmsontheperipheryoftheinternetasonthebackbone.
Thechangeshowsupinemployment.
In2012weidentified5,500peopleworkingatco-locateddatacenters.
In2016wefindthreetimesasmany,15,000employees,andthisnumberexcludesem-ployeesinacquireddatacenterssuchasthoseatworkingatSavvis,whichsince2012becamepartofCenturyLinkandarecountedinitsemploymenttotal.
OurestimateforDataCentersis$6.
1billionand15,233jobs.
Table3.
3.
4:DataCentersEquinix$1,4603,782Akamai$1,4343,727Rackspace$1,0763,999DuPontFabrosTechnology(DFT)$452113CyrusOne$399400CoreSite$333391Internap$318650Quantum$314900Box$2481,123IO$991482016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees30IndistinguishingbetweenHardInfrastructureandSoftInfrastructurewefollowthedistinctiondrawninpriorstudies.
HardInfrastructurereferstoactivitythatisheavyonfixedassets,andSoftInfrastructurereferstosoftwareandservicesbuiltontheHardInfrastructuretomakeittechnologicallyfeasibletoperformcommerceonline.
WeshouldalsoexplainherethedistinctionbetweenthefirmsintheSoftInfrastructurelayeranditsneighbor,theConsumerServicesSupportlayer.
FirmsintheSoftInfrastructurelayeroffergeneral-purposesoftwareandservices,whosevalueappealsatleastinprincipletoallupstreamfirmsregardlessofindustryandwithoutneedingtobecustomizedtoaparticu-larenduser.
FirmsintheConsumerServicesSupportlayerperformservicesthatmustbetailoredtoparticularclients.
Thedistinctionisnotperfect:adagencymediafirms,locat-edintheConsumerServicesSupportlayer,doworkthatmightbeclassifiedintheSoftinfrastructurelayer,whilefirmslikeGartnerandForrester,intheSoftInfrastructurelayerbecausetheysellgeneralreports,alsodocustomwork.
Wemaintainthedistinctiontofollowthechapterstructureofearlierreports.
SoftInfrastructurefirmsareeitherpredominantlyserviceprovidersorsoftwarevendors.
4.
1ServiceProviders4.
1.
1InformationTechnologyConsultingAverysubstantialindustryperformsarangeofservicesthatconnectfirmsintotheHardInfrastructureoftheinternet.
IBISWorld12notesthatbecausecapitalrequirementsarelowandmanyclientsarequitesmall,themajorityoftheseenterprisesaresmall.
Someareone-personoutfitsandmanyareindependent.
Atthesametimethereexistverylargefirmsdesignedtoserveglobalclients,andotherstoserveverylargeU.
S.
militaryandgovernmentagencyclients.
IBISWorldclaimsfromevidencefromtheBureauofLaborStatisticsthatthereareabout440,000firmsandsoleproprietorshipsinthissector,andthatitisoneofthemostresilientelementsoftheU.
S.
economy.
Weanalyzethethreekindsofconsultantinthreeseparatesections.
4.
1.
1.
1EnterpriseITConsultingThesefirmsservethelargestofclients,matchingthefootprintsoftheirofficestotheglob-allydispersedofficesoftheirclients.
Manyoftherelationshipsareseveraldecadesold,originatinglongbeforetheinternetinthecomplexmulti-yearsystemsintegrationproj-ectsrequiredtoimplemententerpriseresourceplanningsoftware.
Whilesuchprojectscontinue,particularlyinfinance,insurance,andsupplychainfirms,innovationssuchascloudservicesandbigdataanalyticsarecreatingdemandforadifferentkindofinter-Chapter4:SoftInfrastructure12IBISWorldIndustryReport54151,ITConsultingintheU.
S.
,September201631netconsultingbythelargestglobalenterprises.
Theinnovationsoftenoriginateinsmallventure-fundedstartupsofferingso-called'pointsolutions'(specificnarrowapplications,)which,iftheythrive,areacquiredbyenterpriseITconsultantstogivethestartupsac-cesstotheenterprisemarket.
AndinsteadoffindingthedecisionmakersinthismarketamongcorporateITdepartments,theyfindthemindivisionssuchasmarketingandop-erationsdepartments.
Inresponse,theseenterpriseITconsultantsrepositionasmarketingandoperationsconsultants.
ThusOracleacquiredDatalogixandBlueKai(smallstartupsinthe2012study)andcreatedtheOracleDataCloudandMarketingCloud,broaden-ingitsconsultingauthoritytoincludemarketinganalyticsandprogrammaticadvertising.
AndIBM—byinternaldevelopmentoftheWatsonanalyticengineandbyacquisitionofdatasupplierssuchastheWeatherChannel—hassoevolvedthenatureofitsconsultingservicesthatitwaslistedbyAdvertisingAgein2015astheworld'sninthlargestadver-tisingagency.
OurestimatesoftheproportionoftherevenuesoftheseconsultingfirmsthatareduetotheinternetrequiredcarefulseparationofinternetITservicesfromservicesthatareindependentoftheinternet.
Wereliedwherepossibleonsegmentinformationreportedin10-Kfilings,and,wherenot,onanalogieswithfirmswherefilingsorinterviewsgaveusabasisforthedistinction.
OurestimateforEnterpriseConsultingis$46.
9billionand210,406jobs.
Table4.
1.
1.
1:EnterpriseConsultingRevenuesandEmploymentIBM$16,49755,000HewlettPackardEnterprise$10,84619,483Oracle$5,34419,617Deloitte$3,39115,225Accenture$3,19736,773Cognizant$1,95234,853CapgeminiNorthAmerica$1,8687,215SAPAmerica$1,2802,778CGI$7406,065FIS$4954,125ComputerSciencesCorporation(CSC)$4593,805Unisys$4223,217DimensionData$4052,2502016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees324.
1.
1.
2PublicSectorandMilitarySystemsIntegrationITconsultants,particularlywheretheinternetisconcerned,oftenspecializeineithermilitaryorgovernmentworkbecausetheyneedsecurityclearancesorbecausetheworkissospecializedastodemandsingle-mindedfocus.
OurestimateforPublicSectorandMilitarySystemsIntegrationis$4.
03billionand17,429jobs.
Weidentifiedthefollowingfirmsasworkingexclusivelyforsuchclients.
Table4.
1.
1.
2:PublicSector&MilitarySystemsIntegration4.
1.
1.
3OtherInternetITConsultingWeenumeratedanumberoffirmsthatdidnotbelongineitheroftheprevioustwocate-gories,butontheevidencethattherewereinexcessof400,000ITconsultants,wechosetoestimatean'allother'category.
IBISWorldestimatedthattherevenuesoftheentireITConsultingindustrywereabout$386.
9billion.
Theirstudyenumeratedthefourlargestintheindustry,whichwere,obviously,alsoindividuallyenumeratedbyus.
WecouldthereforeseetheproportionofITconsultingrevenuethatwedeterminedtobeinternetITconsultingforthosefirms.
Wemadetheassumptionthatinternetconsultingwasahigherproportionofrevenueinenterpriseconsultingfirmsthaninsmaller,particularlysoleproprietorfirms(webdesignbeingseparatelytracked)andarrivedatanestimateofabout$200billionfortheU.
S.
internetconsultingindustry.
Fromthatnumberwesubtractedallindividuallyenumeratedfirms,arrivingatthe"allother"revenuegiveninTable4.
1.
1.
3.
Itisalargenumber,butitiswherealargepartofthedesignofworksuitableformigrationtotheinternetisbeingdone.
CACIInternational$1,4987,960Leidos$9423,600CSRA$5742,430Raytheon$573!
,504ManTechInternationalCorporation$233972NCI$67400MercurySystems$54193LGSInnovations$50130AmericanSystems$452402016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees33OurestimateforOtherInternetITConsultingis$148.
8billionand331,966jobs.
Table4.
1.
1.
3:OtherInternetITConsultingIGATE$1,1433,348SAIC$1,0713,000Syntel$87222,083BTAmericas$5463,500VerintSystems$5191,890Virtusa$378148Datalink$361298Xerox$3302,626ComputerTaskGroup(CTG)$2722,723Collabera$2644,200GeneralDatatech(GDT)$242264GuidewireSoftware$231965AMAXInformationTechnologies$217180Bain$177439VisionIT$170792Datapipe$150378Telos$121521ProgressSoftware$113533Globecomm$106165Workiva$102785Veritas$62100EdgewaterTechnology$47200Ciber$19148Deltek$1780AllOther$141,300282,600FireEyeIncludedin"AllOther"FujitsuUnitedStatesIncludedin"AllOther"2016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees344.
1.
2ResearchersFirmsinthiscategoryproduceindustry-specificresearchandcompetitiveanalysisreportsforcompaniesinthetechnologyindustry,appealingprimarilytothoseholdingpositionssuchasCIO,CTO,andCMO.
Customresearchandconsultingisalsousuallyonofferfromthesefirms,aswellasconferences,workshops,seminars,andwebinarsaimedatkeepingindustrypersonneluptodatewiththelatestdevelopmentsintheirrespectiveareasofendeavor,andtoassistexecutiveswithIT-relateddecisionmaking.
Ouranalysisexaminedthetopthreeprovidersofsuchservices:Gartner,Forrester,andIDCTechnologies.
Gartneristheindustryleader,withapproximately10,000clientor-ganizationsandglobalrevenuesofover$2billion.
Forresterreportedglobalrevenueof$313.
7millionin2015andIDCTechnologiesreported$156.
3million.
Therearemanynewerentrants,suchasCupertino'sConstellationResearch,whichspecializesindisrup-tivetechnologiesandearlyadoptermarkets,andSeattle'sRedMonk,whoseanalystsworkdirectlywithdevelopers.
Anallowanceequaltothesumoftheenumeratedfirmsismade.
OurestimateforResearchersis$422millionand1,722jobs.
HitachiDataSystems(HDS)Includedin"AllOther"IdhasoftIncludedin"AllOther"InforIncludedin"AllOther"IronBowTechnologiesIncludedin"AllOther"KofaxIncludedin"AllOther"MentorGraphicsIncludedin"AllOther"PerficientIncludedin"AllOther"PTCIncludedin"AllOther"SHIInternationalIncludedin"AllOther"TeleTechIncludedin"AllOther"VeevaSystemsIncludedin"AllOther"VeriskAnalyticsIncludedin"AllOther"35Table4.
1.
2:Researchers4.
1.
3DomainRegistryServicesPreviouslytwodiscretelinesofbusiness,domainregistryandwebhostingservicesarenowoftenprovidedbyasingleentity.
Domainregistryisthebusinessofsellingdomainnames,orinternetaddresses,toorganizationsandindividuals,whilewebhostingisthebusinessofhostingwebsitesontheinternet,i.
e.
providingserverspaceandconnectivitysothatindividualwebsitescanbeaccessedontheinternet.
Bothoreitheroftheselinesofbusinessmayalsoincludeemailhosting,websitedesign,anddatahostingservices.
NotablechangesinthisindustryincludethelaunchoffirmssuchasSquarespace,Wee-bly,andWix,whichcombinedomainregistryandhostingwithbuild-your-ownwebsiteservices,providingonestopshoppingforindividualproprietorsandsmalltomediumsizedbusinessessettingupawebpresence.
Domainregistryisestimatedtobea$1.
8billionannualbusinessintheU.
S.
13,withthetoptwofirmsaccountingforover60percentofsectorrevenues.
Webhostingisestimat-edtoberesponsibleforabout$6billioninannualrevenuesintheU.
S.
andisanotherexampleofaverylongtailbusiness,withthetopfourfirmsaccountingforjustunder10percentofsectorrevenuesandover15,000firmsmakinguptheother90percentofthemarket.
14OurestimateforDomainRegistryServicesis$9.
5billionand30,849jobs.
Gartner$130448Forrester$72311IDCTechnologies$9102Allother$2118612016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees13IBISWorldIndustryReport,"WebDomainNameSalesintheU.
S.
,"November201614IBISWorldIndustryReport,"InternetHostingServicesintheU.
S.
,"April201536Table4.
1.
3:DomainRegistryServices4.
2SoftwareVendors4.
2.
1.
AnalyticsSoftwareThissegmentislargelythedomainofcompaniesthatusedtofallunderthemonikerofBusinessIntelligence(BI).
Today'sversionofBItakestheformoftools,applications,dashboards,andvisualizationsoftwarethatanalyzeraw,unstructureddatahousedwith-inanorganization.
Thegoalofthesesystemsistomineforinsightssothatfirmscanact,ratherthanreact,andinturnremaincompetitiveintheirsector.
Overthepastdecadetherewasconsider-ablestartupfirmactivityinthissector,withmultipleacquisitionsfromfirmssuchasIBM,Microsoft,SAS,andSAP.
WeanalyzethelargestU.
S.
firmsinthissectorthathavenotyetbeenacquired.
Ourestimateforthissectoris$941millionand3,596jobs.
DomainRegistryGoDaddy$8052,381Verisign$639611Web.
comGroup$5432,200Neustar$393927WebHostingEnduranceInternationalGroup$3951,3811&1Internet$3501,750Verio$93206Squarespace$83373Hostway$79173LimelightNetworks$72258Wix.
com$53354Bluehost$43175Weebly$760AllOtherWebHosting$6,00020,0002016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees37Table4.
2.
1:AnalyticsSoftware4.
2.
2CustomerRelationshipManagementSoftwareCustomerRelationshipManagement(CRM)softwarehelpsfirmsmanageinteractionsbe-tweenacompanyanditscustomers,aswellasinteractionswithpotentialcustomersandclients.
ThelargestprovidersofCRMsoftwareareSalesforce,SAP,Oracle,andMicrosoft,whotogetheraccountforalmost60percentofthemarket.
Asthesefirmsareanalyzedinothercategoriesofthisreportwedonotincludetheminourcalculationinthissection.
Ourestimateforthissectoris$4.
3billionand14,701jobs.
Table4.
2.
2:CRMSoftwareSplunk$5011,080Cloudera$2151,260TableauSoftware$81375Varonis$71530Qlik$49181Hortonworks$241702016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployeesWest$1,1303,219Sitel$8783,456Sykes$7843,135Synchronoss$5261,724Genesys$3191,313Convergys$215215LivePerson$160664Five9$120375CSGInternational$95414Stefanini$30160Fusion$10262016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees384.
2.
3.
NetworkSecurityThissegmentiscomprisedoffirmsthatcreatesoftwarethatensuressecurityononlinenetworks.
Examplesofsuchsoftwareincludeanti-virussoftware,firewallsoftware,encryptionsoftware,andmalwareandspywareremovalsoftware.
Itisestimatedtobean$11billionmarket,madeupof432firms,ofwhichonefirm,Symantec,claimsjustunder14percentmarketshare.
Thetopthreefirmsaccountfor25percent,andthe'allother'segmentmakesuptheremaining75percent.
15Weassume$3billionofthisrevenueiscontrolledbyfirmsnotintegratedintolargevendorsanalyzedelsewhereinthereport.
Oneofthemoresignificantchangesinthisindustryisthe'bringyourowndevice'environmentinwhichmanyfirmsnowoperate,withemployeesusingtheirownphones,tablets,andlaptopsinadditiontocompany-issueddevicesandmachines.
Thishasledtonewsourcesofthreattoenterprisesecuritysystems,andstartupsandsmallerfirmscomprising75percentofthe'allother'firmsinthiscategoryhaveemergedtorespond.
Inturntherehavebeenacquisitions,aslargeplayersfoundtheycouldmorequicklyin-tegratenewfeaturesandgrowuserbasesbyacquiring,ratherthancreating.
ExamplesofsuchactivityincludeSymantec'sacquisitionsofMessageLabs,OdysseySoftware,Nu-kona,LiveOffice,PasswordBank,andBlueCoatSystems,whichtogetherprovidedthefirmwithofferingsinmessagingsecurity,datacloudsecurity,mobilesecurity,passwordsecurity,andcybersecurity.
Notethatoneofthetopfirmsinthissegment,McAfee,hasbeenfullyintegratedinthesoftwareofferingsofIntelsince2015anditsrevenuesandemployeecountthereforeappearunderIntel'sintherespectivesectionofthisreport.
IntheaftermathofthemassiveOctober2016DDoSattackonDyn,whichtookdowncoreinternetservicessuchasTwitter,Spotify,SoundCloud,andReddit,securityexpertsexpectthesizeandfrequencyofsuchattackstocontinue.
Accordingtoonesuchex-pert,BrianKrebs,"thesizeoftheseDDoSattackshasincreasedsomuchlatelythankslargelytothebroadavailabilityoftoolsforcompromisingandleveragingthecollectivefirepowerofso-calledInternetofThingsdevices—poorlysecuredinternet-basedsecuritycameras,digitalvideorecorders(DVRs)andinternetrouters.
"16Furthermore,thepropor-tionofcorporatecyber-attackssaidtosucceedisnowoneinthree.
17OurestimateforNetworkSecurityis$6.
4billionand17,771jobs.
15IBISWorldIndustryReports,SecuritySoftwarePublishing,U.
S.
,April201616https://krebsonsecurity.
com/2016/10/ddos-on-dyn-impacts-twitter-spotify-reddit/17https://www.
bloomberg.
com/news/articles/2016-11-02/accenture-says-one-third-of-corporate-cyber-attacks-succeed/39Table4.
2.
3:NetworkSecurity4.
2.
4.
VideoSoftwareThoughtheconsumerinternetbeganitslifeprimarilyasaservicetransmittingtextandstillimagesitisnowprimarilyapipedeliveringhighbandwidthvideocontent.
DigitalnativepublisherssuchasMashable,Vox,andMicfeaturevideoasamatterofcourseontheirsitesandapps,whilelegacynews,information,andentertainmentcompaniesoperatingonlineareincreasinglyfindingthattheytooareinthebusinessofdeliveringvideo,anddosousinginternetprotocols.
In2015globalvideotrafficdeliveredontheinternetaccountedfor70percentofallconsumerinternettraffic,andnetworkinghard-warecompanyCiscoestimatesthisnumberwillriseto82percentby2020.
18ConsiderthattheamountofvideouploadedtoYouTube,asingleprovider,hasrocketedfromsixhoursperminutearoundthetimeoftheacquisitionbyGoogletenyearsagoto500hoursperminuteattheendof2015,upfrom300hoursperminuteattheendof2014.
19OtherprovidersofvideobackboneontheinternetincludeconsumerbrandssuchasAmazon,andenterpriseproviderssuchasAkamaiandLimelight,whosereve-nuesandemploymentareincludedinourHardInfrastructurecategory.
Symantec$1,5854,842Fortinet$6362,531BarracudaNetworks$2221,029Gigamon$222357Proofpoint$217986Intralinks$166558Imprivata$119452Qualys$115357Rapid7$83576GlobalSCAPE$2083Allotherindependentfirms$3,0006,0002016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees18http://www.
cisco.
com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/complete-white-paper-c11-481360.
html19http://tubularinsights.
com/hours-minute-uploaded-youtube/40Therefore,ouranalysisofthiscategoryislimitedtotwofirms:BrightcoveandSynacor.
Brightcoveprovidesacloud-basedplatformonwhichbroadcasterssuchasA&EandLifetimehosttheirvideocontent,alongwiththousandsofdigitalpublishersglobally.
Syn-acoralsoworkswithbroadcasterssuchasCNN,aswellaswithcableandbroadbandproviders.
OurestimateforVideoSoftwareis$182millionand612jobs.
Table4.
2.
4:VideoSoftware4.
2.
5AllOtherInternetSoftwareHerewereviewfoursoftwaresectorsthatcollectivelyaccountfor$10billioninrevenueandemploy33,000people.
Thefirstsector,databasetechnology,containssomelargesoftwarefirmsthatarenotyetrelyingontheinternetformorethan25percentoftheirsoftwaredelivery.
ExamplesareRedHat,Linux'sleadingopensourceoperatingsystemsupplier,andTeradata,databasesuppliertosomeofthelargestdatawarehouse-de-pendentcorporations.
Theinternetis,today,notyetamajordeliveryplatformforeither,thoughbothhaveannouncedcloud-basedinitiativesfor2017.
Theinternetsoftwaresectorsuppliestoolsformanagingnetworks,databases,applica-tions,storage,security,andothersystemsacrossmainframe,mobile,andcloudcom-putingenvironments.
ThesectorincludesbroadscopevendorssuchasCATechnologiesandmanyspecialists.
Weusethetermlogisticssoftwaretocoveranumberofvendorswhoautomatethemanagementofmanufacturing,distribution,customerrelationships,andfinancialappli-cations.
Totheextentthatthesevendorsemphasizecloud-basedsubscriptionservices,wecountthemaspartoftheinternetecosystem.
Finallyfirmsinthenetworkaccesssoftwaresectorcreatesoftwarethatenablesremoteworkandcollaboration,mostofwhichridesontheinternet.
OurestimateforAllOtherInternetSoftwareis$10.
6billionand33,639jobs.
Synacor$105284Brightcove$773282016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees41Table4.
2.
5:Database,Internet,Logistics,andNetworkAccessSoftware(AllOtherInternetSoftware)DatabaseSoftwareRedHat$6392,742Teradata$3541,582InterSystems$81245Lionbridge$58618InternetSoftwareCATechnologies$1,2793,495RealPage$4693,640Epicor$4231,000EPAMSystems$4111,016BMCSoftware$330930Broadsoft$279584Demandware$152707JiveSoftware$144530Commvault$69276VASCODataSecurity$2351RealNetworks$1046LogisticsSoftwareCalAmp$233407QAD$111567SciQuest$105510TIBCOSoftware$86278JDASoftware$52465Tangoe$21234SPSCommerce$181182016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees42NetworkAccessSoftwareCitrixSystems$1,7835,163Riverbed$9562,241NETSCOUTSystems$7812,571NETGEAR$780578j2Global$6321,411inContact$195890LogMeIn$18066443CompaniesthatoffercustomizedsolutionstofirmsintheConsumerServiceschapterareanalyzedinthischapter.
Theirproductsareoftenaccompaniedbyprofessionalservicestotailorthesolutionfortheclient.
However,thedistinctionbetweenSoftInfra-structureandConsumerServicesSupportisnotclear-cut.
ITconsultingfirmssuchasForresterandGartnersellgeneralreportsthatappealtoupstreamfirmsregardlessofindustry,butalsooffercustomresearchforclients.
WeanalyzerevenueandemploymentintheConsumerServicesSupportlayeroftheinternetintwoparts:MarketingSupportandOperationSupport.
MarketingSupportreferstothesetoffirmsthathelpfacilitateandpromotetheflowofcommerce,entertain-ment,informationandsocialinteractionovertheinternet.
OperationSupportreferstofirmsthatprovidesolutionsthatenhancecustomerproductivityorenablethemtodeliveragoodorservicetheycouldnototherwisedoefficiently.
ThegrowingimportanceoftheinternetisevidentforfirmsintheConsumerServicesSupportLayer,whichhaveincreas-inglydeliveredtheirsolutionsthroughtheSoftwareasaService(SaaS)model,inwhichthesoftwareisonlyaccessiblethroughthecloud.
Inaddition,anumberoffirmsexistwhosebusinessreliesheavilyontheinternettodownloadapplicationsordataontolo-calservers.
Forinstance,MSCIInc.
deliversinvestmentbenchmarkindexestoinstitution-alinvestmentfirms,insomeinstancesdownloadedonadailybasis,buthostedlocally,oftenforsecurityreasons.
Eveniftheyareusedandstorednatively,digitalproductsandservicescreatedbythesecompaniesrelyontheinternetfordelivery.
Inourpriorstudy,searchenginesandportalsaccountedforover25,000internetemployeesintheConsumerServicesSupportlayer.
ThesearchgiantGoogle'srestructur-ingandre-namingtoAlphabetsoughttoprovidemoretransparencyasthescopeofitsbusinessexpanded.
AlphabetnowresidesintheIntegratorlayeroftheinternetecosys-temasthecompanyhasmadeaforayintobroadbandservices(GoogleFiber),homeautomation(Nest),mobileoperatingsystems(Android),andhasover15mobileappli-cationswithmorethanonebilliondownloads.
20In2015,VerizonacquiredAOLandisintheprocessofcompletingitsacquisitionofYahoo,whichwouldhelptoaddscaletothecompany'ssearchanddigitalcontentproperties.
5.
1MarketingSupportInformationmaywanttobefree,ashasbeensaidsincetheearlydaysofthepersonalcomputerrevolution,buttherealityisthattheinternet'sinformationresourcesareavail-Chapter5:ConsumerServicesSupportLayer20https://en.
wikipedia.
org/wiki/List_of_most_downloaded_Android_applications44ablelargelybecauseofadvertising.
Advertisingmayappearintheformofbanner,display,orsearchads,orasisthecasemorerecently,ascontentmarketingornativeadvertising,whichprovidesinformativeand/orentertainingbrand-supportedcontenttousersinlieuofconventionalads.
Thefirmsandtechnologiesthatmaketheseservicesavailable—advertisingagencies,adnetworksandexchanges,dataandanalyticscompanies,andmeasurementfirms,aswellasself-employedwebprogrammers,designers,andwriters—areexaminedinthissectiononMarketingSupport.
Oneoftheoutcomesofanalwaysonmediaenvironment,accessibletousersthroughphones,tablets,laptops,desktops,anddisplayscreensisafragmentationofbothaudi-encesandattention.
Asaresult,firmsspecializingindeliveryofcontent,services,andtechnologies,haveproliferated.
Furthermore,witheveryclick,swipe,pause,andsharemadevisibleandtrackable,clientshaveanexpectationofgranular,oftenrealtime,informationaboutconsumers,aswellasperformance-basedpricingfromagencies.
Thetablesthatfollowinthissectionwillbreakoutthefirmsthatinhabitboththetradi-tionalterrainofadagenciesandthenewerfirmsinsuchsectorsas'adtech,'whichisshortforadvertisingtechnology,programmaticsellingandbuying,dataandanalytics,andCRM(CustomerRelationshipManagement).
Withtherapidpaceofchangeinthissector,mergerandacquisitionactivityhasbeenoccurringatasubstantialvolume.
Inthefirstsixmonthsof2016,forexample,therewere204dealsvaluedat$6.
8billion,comparedto85dealsvaluedat$2.
1billioninthesameperiodof2015.
21Othernotablechangesinthissectorsincethetimeofourlastreportinclude:Digitaladspendingmeetingtraditionaladspendingforthefirsttime,witheachaccountingfor36percentofbudgets.
22Mobileadspendingnowgrowingatanannualrateof45percent23,settosurpassdesktopadspendingby2017.
24Two-thirdsofdigitaldisplayadspendingisnowtakingplacebywayofautomated,programmaticsystems.
2521http://adage.
com/article/agency-news/digital-properties-dominated-mergers-acquisitions/305397/22https://www.
emarketer.
com/Article/Digital-Ad-Spending-Surpass-TV-Next-Year/101367123https://www.
emarketer.
com/Article/US-Digital-Ad-Spending-Surpass-TV-this-Year/101446924http://fortune.
com/2016/06/20/mobile-internet-ad-spending-desktop/25http://www.
emarketer.
com/Article/More-Than-Two-Thirds-of-U.
S.
-Digital-Display-Ad-Spending-Programmatic/1013789455.
1.
1.
AdvertisingAgencies:FullServiceThefullserviceagencysectorhasremainedmoreorlessstablesincethetimeofourlastreport.
Itisoneofmanyindustrieswithalongtaildistribution,withareported66,291businessesin2015andthetopfouraccountingforabout35percentofrevenues.
26Thesetopfirmsaregenerallyholdingcompanies,housingavarietyofsubsidiarieswithspecialtiesinspecificareas.
Forexample,WPP,oneofthelargestfirmsanalyzedinthissection,istheparenttoGrey,Ogilvy&Mather,Young&Rubicam,J.
WalterThompson,AKQA,andothers.
Similarly,InterpublicistheparentfirmtoCampbellEwald,Deutsch,HillHolliday,McCann,R/GA,andothers.
OurestimateforFullServiceAdvertisingAgenciesis$11.
75billionand52,953jobs.
Table5.
1.
1:AdvertisingAgencies:FullService5.
1.
2Digital&CRMVendorsConcentrationinthissectorofDigital&CRMvendorsissimilartothatinthesectoroffullserviceagencies,withasmallnumberoffirmsaccountingforthebulkofrevenues.
27OurestimateforDigital&CRMVendorsis$13.
02billionand53,733jobs.
OmnicomGroup$2,79713,842PublicisNorthAmerica$2,76311,936WPP$2,4398,851IPG$1,1915,294DentsuAegisNetwork$4423,882AllOtherGeneralFullServiceAgencies$2,1169,1482016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees26IBISWorldReport"AdvertisingAgenciesintheU.
S.
,"September201627IBISWorldReport,"DigitalAdvertisingAgenciesintheU.
S.
,"November201546Table5.
1.
2:Digital&CRMVendorsWhereasthetraditionaladvertisingagencyworldexaminedabovehashistoricallybeenbasedonacombinationofskill,reputation,andrelationships,digitalagenciesbeartheadditionalburdenofhavingtokeepuptodatewiththelatesttechnologicaldevel-Salesforce.
com$4,66713,300AdobeSystems$2,4466,269AllianceData'sEpsilon$1,1693,822Dun&Bradstreet$4591,403Merkle$3032,363ConstantContact$3671,120QuotientTechnology$237553Vivial(formerlyBerryCompany/LocalVox)$188750Marketo$177804Cision(formerlyVocus)$1711,200RhythmOne(formerlyBlinkx)$167316HarteHanks$1521,200HubSpot$155960MaxPointInteractive$140395DialogDirect$1312,250Gainsight$130250Bazaarvoice$128484PegasusSolutions$117699Gyro$117300Tapjoy$100500DefyMedia(AlloyDigital+BreakMediaconsortium)$100370MartinSoftware$73308Sprinklr$50550TechTarget$47612AllOtherDigital&CRM$1,23412,9552016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees47opmentsandconsumerbehaviorsassociatedwithnewsoftware,platforms,anduserexperiences.
Thegrowthofdigitalandmobileadvertisingnotedabove,withtheformernowaccount-ingfor36percentoftotaladbudgetsandthelatterincreasingby45percentyear-over-year,hasledtotheconsiderablegrowthofthisindustrysegment.
Theformidableamountofcontentproductionandconsumptiononline,aswellasitscomplexityacrossdevicesandplatforms,hasledtothegenerationofanunprecedentedamountofdata,sometimesreferredtoas"bigdata.
"Onecharacterizationofbigdataiswiththe3V's:volume,variety,andvelocity.
Makingthisdataintelligibleandusefultoindustryisasub-industrythathasarisen,withfirmscreatingavarietyoftoolsandserviceswithwhichbrandsandagenciescanderiveinsightsfromthesenewlocationsofactivity.
5.
1.
3OnlineAdNetworks&ExchangesToday'smedialandscapeisaworldofcontentdistributedacrossmillionsofwebsites,apps,andplatforms,asopposedtotherelativelysmallnumberofprint,broadcast,andphysicaloutletsavailableforadvertisingplacementinthepre-internetandearlyinternetworlds.
Asawaytomanagethevolumeofinventoryonthesupplyside(publishers),andthedemandside(advertisers),adnetworksandadexchangesemerged.
Theadnetworksfunctionasakindofagentforpublisherinventorythatismadeavailableforsaletoad-vertisers,whiletheadexchangesaretechnologyplatformsthatcreateanopenmarket-placeinwhichadvertisingspaceisbiduponandpurchased,andisgenerallydonebyautomatedsystems,inwhichcaseitisreferredtoas'programmatic'.
OurestimatefortheU.
S.
revenuesandjobsinthiscategoryis$2.
9billionand3,788jobs.
48Table5.
1.
3:OnlineAdNetworks&Exchanges5.
1.
4Measurement&AnalyticsThissegmentismadeupoffirmswhosefocusisthemeasurementofbroadcastanddigitalaudiences.
Someofthesefirms,e.
g.
Nielsen,offersolutionsviasubscriptiononlyforindustrywhileothersmakeconsumerversionswhichareaccessibleonlinewithalimitedfeaturesetavailablewithoutasubscription.
Newdevelopmentsinthissectorincludecross-deviceattributionandrealtimeandcloud-basedtoolsthatmakeperformanceindicatorsavail-ableondemandforagenciesandenterprise.
OurestimateforMeasurement&Analyticsis$778millionand4,272jobs.
Table5.
1.
4:Measurement&AnalyticsMediaMath$800375Exponential$600650iAD28$341195RocketFuel$337755AppLovin$234115PubMatic$130650Tremor$174335YuMe$148259TubeMogul$1464542016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees28Apple,iAd'sowner,discontinuedthesellingandcreationofiAdunitsinmid-2016,cedingtheworktopublishers.
Weassumethattherevenueandemploymentshiftedtodifferententities,butdidnotdisappear.
Nielsen$4152,888comScore$258904Quantcast$1054802016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees495.
2OperationSupport5.
2.
1.
FinancialServicesSupportThecompaniesinthissectionprovidesoftwaretosupportdigitalbankingandtheinvest-mentmanagementindustry.
Robustgrowthinmobilebanking,facilitatedbyexpandingsmartphonepenetrationandenrichedfunctionality,wasakeydriverofthefinancialsup-portserviceslayer.
Ina2011U.
S.
FederalReservesurvey,22percentofmobilephoneuserswithbankaccountshadusedmobilebankingservicesintheprevious12months;inthe2015survey,thisnumberreached43percent.
29SpendingonITserviceshasgrowntoaddresstheneedsofdigitalbankingandpay-mentssuchasenhanceddigitalsecurity.
CompanieslikeFiservandACIWorldwideprovideasuiteofsoftwaretosupportonlinebankingandintermediariesinvolvedintheglobalpaymentsvaluechain.
Theinvestmentindustryisseeingtraditionalbackofficefunctionsmovetoasoftware-as-a-servicemodel.
Someofthelatestintegratedsoftwareplatformshelpwithassetallocationdecisions,reporting,andbillingandcanimproveservicelevelswhileloweringcosts.
OurestimateforFinancialServicesSupportis$2.
2billionand7,569jobs.
Table5.
2.
1:FinancialServicesSupport5.
2.
2E-learningSupportE-learningSupportreferstoasegmentofcompaniesthatallowschoolstodeliverorman-agelearningthroughtheinternet.
Weestimateinternetrevenueof$0.
8billionand4,656individualsemployedinthesector,whichiscomprisedprimarilyoftwotypesofcompa-nies:OnlineProgramManagement(OPM)andLearningManagementSystems(LMS).
Fiserv$7413,102MSCI$519964Envestnet$388999ACIWorldwide$232897BottomlineTechnologies$193876SS&CTechnologies$1717312016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees29https://www.
federalreserve.
gov/econresdata/consumers-and-mobile-financial-services-report-201603.
pdf50OPMssuchas2U,HotChalk,andTheLearningHouseenableschools,oftenpost-secondaryinstitutions,toofferinstructor-ledclassesonline.
Throughaweb-basedplatform,instructorscangivelectures,leadgroupdiscussions,anddisplaylearningmaterialstoavirtualaudi-ence.
OPMsofferbundledservices—whichincludeacoreplatform,technologysupport,contentdevelopmentandmarketingtoprospectivestudents—orunbundledservicestohelpinstitutionsmeeteachprogram'sspecifications.
Virtuallearningallowsuniversitiestoincreaseenrollmentprofitablywithoutthecostburdensimposedbyphysicalinfrastructureandotheroverhead;ithasgainedwideracceptanceandtrustamongstemployersoverthequalityofonlineprograms.
ArecentsurveybyEduventures,aneducation-focusedresearchfirm,foundthat350U.
S.
universitiesareengagedinOPMpartnershipsintheU.
S.
andthisnumbershouldcontinuetogrowsupportedbyuniversities'pursuitofadditionalstreamsofincome.
30AnLMSisacloud-basedsoftwareapplicationthatprovidesacentralizedplatformtostoreandaccesseducationmaterials,deliverassessments,andtrackastudent'slearningprog-ress.
ALMSalsosimplifiesadministrativefunctionslikecourseregistrationandattendancemetrics,andcancaptureandanalyzedataontheefficacyofdigitalcontentanduseren-gagement.
Blackboard,anincumbentfirmofferingaleadingLMS,expandeditsworkforceby30percentsinceourlaststudy.
NewentrantssuchasInstructure—withitsbrandedLMS'Canvas'—aregainingmomentumamongreputableinstitutionsofhigherlearning.
HarvardUniversity,forexample,recentlytransitionedfromitsinternallydeveloped,decadeold,LMS'iSites'toa'Canvas'solution.
Inadditiontothepost-secondaryspace,therearecountlessinnovativeK-12LMSsolutions,includingfunctionstoengageparentsontheplatform,whichhaveattractedlargeuserbaseswheremonetizationisnotastrategicobjective.
ClassDojo,aleadingK-12mobileLMSapp,wasfoundedin2011ishelpingtoimproveteacher,studentandparentcommunicationandisactivelyusedintwooutofeverythreeU.
S.
schools.
31OurestimateforE-learningSupportis$799millionand4,656jobs.
Table5.
2.
2:E-learningSupportBlackboard$5342,3402U$150904Instructure$69725HotChalk$12400LearningHouse$342872016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees30http://monitor.
icef.
com/2015/10/private-partners-helping-to-drive-growth-of-us-higher-education-online/31https://www.
classdojo.
com/press/515.
2.
3TravelServicesSupportEmploymentinTravelServicesSupportisprimarilycomprisedofindividualswhoworkforGlobalDistributionService(GDS)companiessuchasSabreandTravelport.
GDSfacilitatestravelcommercebyconnectingtravelproviders,suchasairlinesandhotelchains,withofflinetravelandonlineagencies,suchasPricelineandExpedia,andothertravelbuyersbydisplayinginventory,prices,andavailabilityfromtravelsuppli-ersallowingconsumerstopurchasethatcontentthroughthemarketplace.
GrowthoftheGDSindustryisbolsteredbytailwindsfromrisingairtravel—theInternationalAirTransportAssociation,atradeassociationoftheworld'sairlines,forecastsglobalairlinepassengergrowthof3.
7percentperannumoverthenext20yearsandNorthAmericatogrow2.
8percentoverasimilartimeframe32—buttheyalsofaceheadwindsfromtheriseofmeta-searchtravelwebsitesandairlinespromotingdirectbookings.
OurestimateforTravelServicesSupportis$1.
2billionand1,906jobs.
Table5.
2.
3:TravelServicesSupport5.
2.
4HealthcareInformationSolutionsNationalhealthcarespendingisalargesectoroftheU.
S.
economy,accountingfor$3trillioninannualspending,andhealthcareITspendingcurrentlyrepresentstwopercentofthissum.
33Electronichealthrecords(EHR)formthebackboneofhealthcareinformationsolutions.
EHRsaredigitalversionsofahealthcarepatient'spapercharts.
Federallegislation—aspartoftheAmericanRecoveryandReinvestmentActof2009—createdincentivestohelpdoawaywithinefficientpaper-basedrecordkeepingpracticesandsupportawaveofEHRplatformadoptionsinceourlaststudy.
Majorbeneficiariesincludethecompa-niesCerner,Nuance,andEpic,whotogethercompriseabout85percentofthehospitalEHRmarket.
Ascomfortwithstoringpersonalinformationincloudisgrowing,manyofTravelportLimited$755315SabreCorporation$4551,399DeltaTechnologies$151922016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees32http://www.
iata.
org/pressroom/pr/Pages/2016-10-18-02.
aspx33https://www.
cms.
gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/nhe-fact-sheet.
html52thecompaniesintheEHRindustryhavebeenshiftingtheirsoftwareplatformsfromonpremiseontothecloud,helpingtofacilitatetheaggregationofdatafromvarioussourc-esontoasingleplatform.
Thesecompanieshavealsobeendevelopingtools,inpartner-shipwithhospitals,toprovideawindowintotheoverallhealthofthepopulation,withtheultimategoalofidentifyingcommonrisksandtakingproactivestepstointervene.
Thistrajectoryisanaturalevolutionindigitalinnovation:asmorepersonalhealthdataisrecordedandstored,advanceddataanalyticscanbeleveragedforinsightstoim-provehealthcareoutcomesandlowercosts.
Analyticsarealsoutilizedtospeedupdrugdevelopment.
MedidataSolutionsInc.
,forinstance,offersacloudplatformwithanalyti-calcapabilitiestostreamlineclinicalresearchandthedrugdevelopmentprocess.
OurestimateforHealthCareInformationSolutionsis$2.
4billionand11,783jobs.
Table5.
2.
4:HealthCareInformationSolutions5.
2.
5ShippingAlargepackageshippingindustryhasdevelopedtosupportthesteadygrowthofe-commerce,whichhasaveragedabout15percentforthepastsixyears.
34Weestimatethatin2015about11.
6billionpackagesweremovedintheU.
S.
,basedondatapub-lishedbyFedEx,USPSandUPSandextrapolationtoothershippersbasedonmarketshare.
Publicstatementsoftheseshipperssuggestthatabout4.
6billionofthesepack-agesweree-commercepackages,andofthemweinfer1.
26billionwereshippedonbehalfofAmazon.
OftheAmazonpackagesUPSandFedExmovedabout400million,withtherestcarriedbyacombinationofAmazon'sownresources,USPS,andothercarriers.
Cerner$1,1305,619NuanceCommunications$4602,293EpicSystems$3962,090MedidataSolutions$298967CPSI$59488Meditech$413262016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees34https://www.
internetretailer.
com/2016/02/17/us-e-commerce-grows-146-201553AstherearethousandsofshippingandlogisticsprovidersintheU.
S.
forouranaly-sisweperformedarollupoffirmsintheU.
S.
thatbenefitfromthedeliverybusinessenabledbye-commerceandarrivedatrevenueof$30.
3billionandanemploymentcontributionof353,315.
Table5.
2.
5:Shipping5.
2.
6E-governmentSupportCompaniesinthissegmentprovidetechnologyandconnectivityservicestolocal,state,andfederalgovernment,enablingeverythingfromofficialgovernmentwebsites(e.
g.
whitehouse.
gov,data.
gov)to,inthecaseofSilverSpringNetworks,thecloudinfrastruc-tureonwhichpublicutilitiesoperateinsomelocales.
Insuchascenarioservicesarecalled'networkasaplatform',andconstituteyetanotherexampleofthemodularizingofinfrastructureweseethroughouttheongoingevolutionoftheinterneteconomy.
AprovidersuchasTylerTechnologies,bycontrast,focusesonsoftwaresolutionsforlocalgovernmentandthepubliceducationsystem,whileNICInc.
isanITserviceproviderto3,500stateandlocalgovernmentagencies.
OurestimateforE-governmentSupportis$829millionand2,105jobs.
Table5.
2.
6:E-governmentSupportShippersdeliveringonlinepurchases$30,284353,3152016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployeesSilverSpringNetworks$426567NIC$292859TylerTechnologies$1116792016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees545.
2.
7WebEnablingServicesWelistherethesubstantialnumberofself-employedwebdesigners,developers,andcontentcreatorswhoseworkisnotcapturedelsewhereinourstudy.
Ourestimateforthissegmentis$17.
9billionand182,284jobs.
WeuseddatafromtheU.
S.
BureauofLaborStatistics(BLS)toassistusinthefollowingprocess:Forappdevelopersformobiledeviceswefollowedthe2012study'smethodology.
Inour2012reportwehadreliedonastudybyMandel35andnomorerecentsourcecouldbefound.
Therefore,weprojectedthatappemploymentwouldgrowfrom2012to2106attherateatwhichitwasgrowingin2012.
Forwebcontentcreators,webeganwitha2015BLScountof189,840writersandeditors.
Weattributed20percenttowebwriting,assumedthat62percentwereself-em-ployed,andtook22percentaspart-time36whocouldbedisregarded.
Multiplyingtheseproportionsyielded23,540people.
Butwelist10,000writersoncontentsitesthathirefreelancewritersandbloggers,sonetincrementalemploymentis13,540.
Forwebcomputerprogrammers,webeginwiththeBLS2015totalforsoftwaredevel-opersof747,730.
Theproportionwhoworkonwebsoftwaredevelopmentandareself-employedisanchoredtoourestimateofwebdeveloperssincefreelancedevelopmentrequiresprogramming,andisdeflatedto7.
5percentoftheBLSnumbertoallowfordupli-cationofskills.
Thisamountstoasixpercentcompoundedgrowthratesince2010.
Forwebdevelopersandgraphicdesigners,wenotethattheBLSreportsadeclineintherelevantemploymentcategories.
Thismaybebecausewebhostingcompaniesnowoffersimpledo-it-yourselfwebcreationtools(seeSection4.
1.
3.
)Workingfromthe2015BLSemploymenttotals,ourassumptionsarethat20percentofthegraphicdesignersarewebdesigners,35percentofU.
S.
webdevelopers/computerprogrammersareself-employed,andthat14percentinthecategoriesofwebdevelopersanddesignersworkpart-time.
37Forwritersourassumptionsarethat20percentofthoseidentifiedbyBLSdatawritefortheweb,that62percentareself-employed,andthat22percentarepart-timeworkers.
38Forallcategorieswemaketheassumptionthattheearningsfrompart-timeemploymentarenotmaterialandthereforedonotincludetheminourestimate.
Thetablebelowbreaksoutthefullestimatesforrevenuesandemployment.
Weestimaterevenuesforeachoccupationbymultiplyingfully-burdenedlaborcostderivedfromBLSaveragewagesbyemployment.
35Mandel,M.
(2012)"WheretheJobsAre:TheAppEconomy,"TechNet,SouthMountainEconomicsLLC.
36Thefollowingsitedescribesacomprehensivesurvey,butwithoutattribution:http://www.
studentscholarships.
org/salary_ca/102/web_designers_and_developers.
php37http://www.
studentscholarships.
org/38http://www.
studentscholarships.
org/55Table5.
2.
7:WebEnablingServices5.
3GeneralEnterpriseBeginningwiththe2008study,anallowancewasmadeforgeneralenterpriseomittedemployment.
Bythiswemeantthattherewouldbepeopleworkinginlargecorpora-tions,non-profitorganizations,andgovernmentagencies,whoowedtheirjobstothein-ternetbutwhoseemployerswerenotconspicuousmembersoftheinternetecology.
Weperformeddetailedenumerationonabout440firmsandabout20rolled-upfirmgroupsandself-employedgroups,butthereare29millionfirmsintheUnitedStates,ofwhichsixmillionarelargeenoughtohaveapayroll.
Granted,manyhavenointernetemploy-eesandothersmakeinternetdutiesapart-timeresponsibilityofoneperson,butin2008wemadeanallowanceforgeneralenterpriseomittedemploymentat100,000people,andin2012at170,000people.
Giventhatomittedemploymentshouldbeproportion-altogrowthinoverallinternetecosystememployment,whichhadgrownaboutfourfoldsince2008,weestimatedomittedemploymenttobe390,000in2016.
Table5-3:GeneralEnterpriseActivityAppdevelopersformobiledevicesascontractors$6,07761,897Webcomputerprogrammers$5,68056,267Webdevelopersandgraphicdesigners$4,41250,580Webcontentcreators(writers)$1,78613,5402016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployeesGeneralEnterpriseActivity$97,500390,0002016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees566.
1ContentSitesThoughtheinternetcouldnotexistwithouttheinfrastructureandmachinerymappedintheearlierchaptersofthereport,itiscontentanduserexperiencethatdefinetheinter-netforitshundredsofmillionsofusersintheU.
S.
Theseconsumertouchpointsincludewebsites,apps,andmessagingservices.
Thecon-tentrangesfromnewstoinformation,music,audio,photos,andvideotoservices,andisaccessedonphones,tablets,laptops,desktops,andlivingroomscreens.
Sometouchpointstaketheformofplatformsthatcreatemarketplacesforarangeofservicesfromdatingandpeer-to-peerlendingtosellingbabyclothes.
Theymaybesitesthatallowcollectiveknowledgetobeamassed,suchasWikipediaandQuora,sitessuchasDai-lyStrength.
orgthatallowpersonalissuesandburdenstobesharedanonymously,sitessuchasIdeaConnectionthatencouragesindividualstosolveproblemsinexchangeforfinancialreward,orSeeClickFix,whichallowsanyoneinacommunitytoreportanissueofcivicconcernforremediation.
Somearefor-profitenterprises,somearenot-for-profit.
Somecarryadvertising,whileothersdonot.
Formsoffundingrunthegamutofone-timedonationstosubscriptionstoaffiliatemarketingpaymentstoadvertisingrevenuesharestolaborsoflove.
Itis,afterall,theinternet,wherethebusinessmodelorformofsuste-nanceoftencomesaftertheactofpublishing,notbefore.
Theinternetasaconsumer-facingexperienceisnowover20yearsold.
Forthegener-ationborninthe1990sandnowenteringyoungadulthood,therehasneverbeenaworldwithouttheinternet.
It'snotsurprisingthereforetoseealongarcofevolutioninthesourcesofpublisherrevenue,aswellasshortcyclesofdisruption.
In2016,afterslowbutrelentlessgrowththatbeganinthe1990s,advertisingspendingondigitalmediareachedthelevelofspendingonTV,eachmediumreceiving38percentoftotalspending.
392016alsomarkedfive-yearsduringwhichtelevisionviewingof18to24yearoldsdecreasedbymorethanninehoursperweek.
Thisdropintraditionalmediaconsumptionrepresentsapproximately40percentofthisdemographic'stelevisionview-ingtimemigratingtodigitalplatformsortootheractivities.
40Thesectorofonlinenewsandinformationcontinuestoroil,thoughover20yearshaveelapsedsincetheintroductionoftheconsumerinternet.
Whileononehandoftheequa-tionwehaveacornucopiaofcontent,services,andtoolsavailedtousinlargepartbytheadvertising-supportedinternet,ontheothersideoftheequationwehavethe7039https://www.
emarketer.
com/Article/TV-Digital-Dead-Heat-U.
S.
-Media-Dollars/1014552/40http://www.
marketingcharts.
com/television/are-young-people-watching-less-tv-24817/41http://www.
emarketer.
com/Article/U.
S.
-Ad-Blocking-Jump-by-Double-Digits-This-Year/1014111Chapter6:ConsumerServices57millionAmericansusingadblockersin2016,a34percentyear-over-yearincreasethatisprojectedtoincreaseanother24percentin2017.
41Buttheinternetisnothingifnotendlesslyadaptive,andinthewakeofthevigorousuptakeofadblockingcomesnewformats,suchascontentmarketing,brandedcontent,andnativeadvertising,whichprovideworkaroundstothetraditionaladformatsunderattackfromthearmiesofblockers.
Sincethetimeofthelastreportsomeofthemostnotablechangestotheonlinecontentlandscapehavebeen:Thefurthermaturingofsocialmedia:Facebookgrowsto1.
7billionusers,Twittertoover300millionusers,andSnapchat—whichdidn'texistatthetimeofthelaststudy—nowreceivingover10billionviewsperday.
Inmusictheshiftfromdownloadingtostreamingisalmostcomplete;Appletalkingaboutdiscontinuingdownloadsaltogether;Spotifywhichhadatrivialamountofrevenueatthetimeorourlastreportisnowthedominantproviderofmusicstream-ing,withapproximately100millionusersofwhichcloseto40millionarepayingsubscribers.
ThemainstreamingofUGC(User-GeneratedContent)fromanactivityviewedlargelyasanoveltytoaformatthatoccupieseverythingfromcommentssectionsinonlinenewspaperstocollectiveknowledgesitessuchasQuora,Reddit,andWikipedia,tothevideosof'vloggers'reviewingitemsfromcosmeticstocarstosneakers.
Anindustrializationofindividualcontentcreators,agentsandnetworksspecializingininfluencermarketingemerging,resultingin,forexample,themorethan$1.
5billionand35,000-plusfulltimeequivalentjobswehaveestimatedforU.
S.
YouTubecreators.
6.
1.
1News&InformationThissegmentiscomprisedofthefollowingtypesofcompaniesandcontentwhichfallunderthecategoryofonlinenewsandinformation:ThedigitalactivitiesofsuchmediaindustrycornerstonesasBloomberg,Gannett,andTheNewYorkTimesCompany.
ThenotforprofitentitytheWikimediaFoundation,whichhousestheuser-generatedencyclopediaWikipedia,recipientofabouteightbillionmonthlypageviews42andtherelativelysmallassociatedrevenueof$26.
5million.
42http://reportcard.
wmflabs.
org58Thenative-to-digitalfirmsthatarenewwieldersofinfluenceinthislandscape.
NativetodigitalfirmsexaminedincludeXOgroup(publishersofpopularwedding,home,andlifestylesitessuchasTheKnot,TheNest,andTheBump),realestatedata-baseZillow,andSheKnowsMedia.
SheKnowsisafirmnewtoourstudy,andisthepar-entcompanyofafamilyofwomen'slifestylebrandssuchasStyleCaster.
com,BlogHer.
com,andDailyMakeover.
com.
Ithasemergedasthetopdigitalmediacompanyinitssector,claiming80millionmonthlyuniquevisitorsandcloseto200millionfansacrosssocialmediaplatforms.
43Whilecontentmarketingandnativeadvertisingmovedtotheforebetweenourlaststudyandthisone,otherpublishingmodels,deemedinnovativebysomeandcontroversialbyothers,retreated.
Inthiscategorywefindthefollowingcompanies,whoserevenuenumbersfellconsiderablybetween2011and2015and/orbusinessesfoldedorwereforcedtosellassetsduetolegalproceedings.
DemandMedia–Roundlycriticizedforits'contentfarm'model,withitslowcostas-semblylineforcontentthatmatchedthemostpopulartermsandphrasessearched.
Itsrevenueshalvedbetweenourtwostudies,andthecompanyhasdivesteditselfofitsCracked.
comandTrails.
combrandsaswellasitsdomainnameservices.
GawkerMedia–Ahighprofiledefamationlawsuitledthecompanytoliquidateitsassetsin2016,withUnivisionCommunicationsacquiringitsonlinecontentbrands(Deadspin,Gizmodo,Jalopnik,Jezebel,Kotaku,andLifehacker)atauction.
GlamMedia–Acontentaggregatorspecializinginlifestylecontentonline,withthird-partysitesfocusedonsuchtopicsasbeauty,health,food,fashion,entertain-ment,andparenting,thecompany,whichcametobeknownasModeMediaandwashailedasa'unicorn',i.
e.
astartupwithabilliondollar-plusvaluation,abruptlyceasedoperationsinfall2016.
OurestimateforNews&Informationis$18.
2billionand47,777jobs.
43http://www.
brandchannel.
com/2015/08/06/5-questions-sheknows/59Table6.
1.
1:News&InformationTable6.
1.
1:News&Information6.
1.
2Multi-genreContentThiscategoryisdistinctfromthepreviouscategoryofNews&Informationasthefirmsanalyzedherepublishand/ordistributecontentnotmerelyacrossawidevarietyofplatformsbutacrossawidevarietyoftopicareas.
Bloomberg$5,8597,350ThomsonReuters$4,10817,494RELXGroup(formerlyReedElsevier)$3,4408,336Gannett$6774,301Zillow$6452,204WebMD$6361,740Ancestry.
com$5491,074TheNewYorkTimesCompany(Digital/OnlineProperties)$434961Bankrate$368512TrueCar$260574Answers.
com$245622EverydayHealth$232700UnitedOnline$155145Autobytel$133200DemandMedia$126350XOGroup$104444SheKnowsMedia$100250ModeMedia(formerlyGlamMedia)$100240WikimediaFoundation$272802016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees60OurestimateforMulti-genreContentis$19.
6billionand34,391jobs.
Table6.
1.
2:Multi-genreContentAsintheNews&Informationcategoryaboveweonceagainfindameetingofoldworldandnewworldmediacompanieshere,withlegacymediaorganizationssuchasHearstacquiringdigitalupstartsComplexMedia,andnewer,digitalnativebrands,successfullycarvingoutterritoryalongsidethenamebrandsofbroadcastingandpublishing.
Thefollowingcompaniesfromtheanalogworldarenotablefortheiradditionofdigitalentitiestobroadcastbrandsand/orthecreationoracquisitionofstandalonedigitalbrands:ScrippsInteractive:e.
g.
HGTV,FoodNetwork,DIYNetwork,CookingChannel.
TribuneMediaCompany:e.
g.
TribuneStudios,TribuneDigitalVentures,andtherecentlyrenamednewspaperpublishingarm,Tronc.
TimeWarner$3,3672,969Viacom$3,3092,429Hearst$2,8133,273IAC/InterActiveCorp$2,3763,700SonyUSA$1,8684,840LibertyInteractive$1,2462,754NewsCorporation$9772,970DiscoveryCommunications$868941TribuneMediaCompany$6501,750AdvancePublications$6302,905Disney$5872,069Meredith$316765CBSInteractive$2301,894ScrippsNetworksInteractive$137158Univision$94139BuzzFeed$85650ViceMedia$501852016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees61DiscoveryCommunications:e.
g.
DiscoveryChannel,TLC,AnimalPlanet,equityinterestinOWN(OprahWinfreyNetwork).
NewsCorp.
:e.
g.
TheWallStreetJournal,HarperCollinsPublishing,Realtor.
com,socialmediacontentagencyStoryful.
TimeWarner:e.
g.
(viaTurnerBroadcasting)cablechannelsCNN,TBS,andTNT,pay-TVchannelsHBOandCinemax,WarnerBros.
TV,film,homevideo,andgames.
HearstCorporation:e.
g.
majordailyandweeklyU.
S.
newspapers,300magazinetitlesincludingCosmopolitanandEsquire,HearstTelevision,HearstEntertainment&SyndicationUnit,andstakesincablenetworksA&EandESPN.
Existingalongsidetheexpandingmediaportfoliosofthefirmsnamedabovearecompa-niessuchasViceMedia,whichenteredthemedialandscapefromtheextremesofthemargins,andnowhasequitystakesheldbyDisneyandRupertMurdoch'smediaem-pire.
Vice'sin-your-face,outlawbrandofjournalismandentertainment,deliveredacrossdigitalandanalogmediaproperties,hastranslatedtoannualrevenuesapproaching$1billion.
AnothercompanynoteworthyinthiscategoryisBuzzFeed—adigitalbrandoncesyn-onymouswiththedisparagingterm'clickbait'thathasbecomearapidlygrowingandprofitablemediaentity—enjoyingsevenbillionmonthlyviewsasofearly2016,44cov-eringeverythingfromlightfaresuchasentertainmentgossip,food,andlifestylestoriestoserioustopicssuchaspolitics.
Inthecaseofthelatter,theagilityandabilityoftheBuzzFeedpoliticsteamshonetosuchanextentthatCNNhiredthemawayforitsownCampaign2016digitalcoverage.
45Lookingatthedatafromthestudiesourteamhasconductedin2008,2012,andnow2016,itisevidentthatinthedigitaldomain—unlikethebricksandmortarworldofhighbarrierstoentry—enterprisescangofrombarelybeingontheradarinonereporttoadominantforceinthenext.
Theinverseisalsotrue;categoryleaderscanbecomeal-so-ransinthecourseoffouryears,andentirelynewcategoriescanemerge,blindsidingincumbentswhosegazeswerefixedelsewhere.
Aninnovationthathasemergedbetweenthe2012reportandthisstudyisthedigitalincarnationofsponsoredandbrandedcontent.
Theconceptitself—ofbrandsaligningthemselveswithprogrammingnotabouttheirbrandbutaboutgeneralinterestcontentmeanttoattractspecificaudiencesegments—isnotnew,harkingbacktothedaysofTV'ssoapoperasandprogramssuchasMutualofOmaha'sWildKingdom.
Whatis44http://www.
adweek.
com/news/technology/buzzfeed-now-getting-7-billion-content-views-month-171182/45https://www.
washingtonpost.
com/blogs/erik-wemple/wp/2016/10/03/cnn-hires-buzzfeed-scoopsters-for-its-digital-politics-coverage/62differentthistimeisthemechanicalandtechnologicalendofthings,suchastheabilityofthiskindofcontentto:Bypassadblockers(asthecontentisnotidentifiedbythesystemasanadandisthereforedeliveredtotheuser/viewer/reader)Takeonthelookandfeelofthepublishersitewherethecontentappears,hencetheterm'nativeadvertising',inwhichthesponsoredcontentappearstobenative—orfullyintegratedwith—thepublishersiteonwhichitappears.
Sponsoredcontentcanappearwhereaudiencesalreadyare—themostpopularsocialplatforms,blogs,andsites,andoffercontentthatspeakstotheirinterests,asopposedtomerelypromotingaproductorservice.
Targetingtechnologiescandynamicallymatchthecontentmostlikelytoappealtothereader/viewerwithsuchdatapointsasdemographics,psychographics,andpurchasehistory.
Thesetechniquescombinenativeadvertisingwithprogrammatictechnologies.
Sopopularisthismorecustomer-centricapproachtomarketthat31percentofpublish-ersreportthattheynowhavetheirownin-housefacilitiesfortheproductionofnativeadvertisingandbranded/sponsoredcontent46androbustdigitalpublishingbusinessmodelshavebeenbuiltuponthestrategy,withthemajorityofBuzzFeed'srevenuescomingfromnativeadvertising,asignificantportionofVice'sstemmingfromthesame,andpublicationssuchasTheAtlanticreporting75percentoffitsadrevenuecomingfromsponsoredcontent,andSlate'srevenuebreakdownnowat50percentsponsoredcontentand50percentbanneranddisplayadvertising.
472015wasalsothefirstyearpodcasting—atechnologythathasbeenaroundforoveradecade—appearedatNewYork'sannualadvertisingUpfronts.
Theinternet-enableddigitalaudiosectorwasbuoyedlargelybytwofactors,whichdevelopedinparallel:TherunawaysuccessofSerial,atruecrimepodcastdownloaded100milliontimesTheearlystagesofindustrymaturation,evidencedbytheappearanceofpodcastingnetworkssuchasPanoply,Gimlet,Earwolf,andnetworksandfirmsspecializinginpodcastadvertisingplacementsuchasMidrollandPodtrac.
Relativetootheradvertisingspendsthedollarsbeingdirectedtopodcastpodcastingad-vertisingareinfinitesimal—e.
g.
$34millioncomparedtoradio's$17billionandtelevi-sion's$66billion48—butasthedigitaltransformationofthecontentindustriescontinues46http://www.
foliomag.
com/majority-of-publishers-use-their-own-editorial-staffs-to-produce-native-ads/47http://www.
nytimes.
com/2016/07/25/business/sponsored-content-takes-larger-role-in-media-companies.
html48http://www.
wsj.
com/articles/podcasts-face-advertising-hurdles-145574549263toevolveweexpecttheformernumbertogrowandthelatternumberstoretreatintheyearstocomeacrossdigitalaudioservicessuchaspodcasts,streamingmusicservices,andthedigitalliveandon-demandservicesoftraditionalaudiobroadcasters.
6.
1.
3SpecializedResearch&UserGeneratedContentInthissectionwereportonservicesthataggregateconsumerreviews,businessre-views,andotherratingsservices.
WhileGoogledominatesgeneralpurposesearchandresearchonline,andoffersreviews,ithasnotdisplacedthedemandforspecializedresearchsites.
Theinnovationthatcharacterizesmostfirmsinthissectorismergingoftrustwithawide-lydispersedcrowdwithwhomyoudon'thaveapersonalrelationship,andtheresultingeconomyofreputations.
ArecentstudysuggeststhattwothirdsofU.
S.
internetuserstrustbusinesseswithpositiveonlinereviewsmorethanbusinesseswithnoreviews,49sotheeconomicimpactofsuchsitesissignificant.
Thelargestofthespecializedresearchfirmsareanalyzedindividually,andanestimateismadeofallothers.
NotethatouranalysisofcrowdsourcedtravelreviewsiteTripAdvi-sorisnotincludedhereasitappearsinoursectionononlinetravel.
OurestimateforSpecializedResearchandUGCis$933millionand4,032jobs.
Table6.
1.
3:SpecializedResearch&UserGeneratedContent(UGC)6.
1.
4OnlineMusicServicesMusicisoftenreferredtoasthecanaryinthecoalminefordigitalbusinesses,thefirsttobesentintothedarkness,withothersectorsabletolookoveritsshoulderandobservetheoutcome,howeverdire.
Atpresentthere'sgoodnewsandbadnewsforthisindustry.
WhileU.
S.
musicbusinessrevenueshaveplungedbyover70percentbetween1999and201550thebetternewsisthattheindustryreporteditsstrongestgrowthsincethepivotalNapstereraofthelate1990s.
51RevenuesfromdigitalmusiccontributedtheYelp$4951,998Angie'sList$3441,730Hoover's$943042016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees49http://www.
emarketer.
com/Article/Web-Users-Put-More-Stock-Consumer-Reviews/101292950https://redef.
com/original/less-money-mo-music-lots-of-problems-the-past-present-and-future-of-the-music-biz51https://www.
riaa.
com/reports/2016-mid-year-riaa-shipment-and-revenue-statistics/64majorityofthisincome,asforthefirsttimestreamingtoppedpaiddownloads,asaleschannelnowindecline.
52Thistrendwasbuoyedbytwofactors:thegrowingpopular-ityofmusicstreamingservicesavailableinad-supportedfreemiumandad-freepaidformats,andtheconvenienceofmobilelisteningviaappandsmartphone,withmobilelisteningrepresenting74percentofdigitalaudioconsumptionintheU.
S.
in2015.
53OurestimateforOnlineMusicServicesis$2.
5billionand3,163jobs.
Table6.
1.
4:OnlineMusicServicesThemajorfirmsexaminedinthissectorwerePandora,Spotify,Rhapsody,andiHeart-Media(formerlyClearChannel).
TherevenuesofcompetitorsGoogleMusic,AppleMusic(launchedin2015),andAmazon'smusicservicearecapturedinouranalysesofthosefirms,whichappearintherelevantcategoriesofthisreport.
CompetitorssuchasSlackerand8trackswereshowntohavenegligiblerevenues.
Thebigstoryinthedigitalmusicspacesincethetimeofourlaststudy,publishedin2012,hasclearlybeenSpotify,surgingfrom$3millioninU.
S.
revenuefouryearsagotowellover$1billionin2015.
CompetitorPandoraalmostquadrupleditsU.
S.
reve-nuesbetween2011and2015,from$274millionto$931million,andpickedupad-ditionaluserswithitsacquisitionofstreamingserviceRdio,afteritfiledforbankruptcyinlate201554.
Rhapsody,oneofthepioneersinthisspacewithadecadeofoperatinghistoryinonlinemusic,reported$220millioninglobalrevenuein2015andap-proached3.
5millionsubscribers.
55Rhapsodyoffersnofreetierforitsservice,nordoestheJayZ-ownedTidal,whichreportedthreemillionsubscribers,ofwhich45percentsubscribedtothe$19.
99/monthtierandtheother55percenttothe$9.
99/monthtier,forglobalrevenuesofapproximately$43.
5millionin2015.
56NotethatthecompanyisSpotify$1,240157PandoraMedia$9311,997iHeartMedia$312935Rhapsody$75742016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees52http://ifpi.
org/news/IFPI-GLOBAL-MUSIC-REPORT-201653viaeMarketer,"U.
S.
DigitalAudioListeningShare,Desktopvs.
Mobile,"publishedApril28,2015.
54http://www.
billboard.
com/articles/business/7519014/rdio-bankruptcy-story-how-it-happened-failing-streaming-service55http://www.
billboard.
com/articles/business/6806086/rhapsody-2015-global-subscribers-growth-streaming56http://www.
billboard.
com/articles/business/7317891/tidal-one-year-anniversary-by-the-numbers65headquarteredinEurope,withsmallnumbersforU.
S.
revenueandemploymentandisthereforenotbrokenoutseparatelyinournumericalanalysis.
Particularlynotableinthe'allother'categoryisSoundCloud.
Launchedin2008,thesiteenjoysauserbaseof175millionusers,andisafavoriteofbothfansandmusiciansforitsdistinctlynon-commercialslant.
Theventurecapital-backedcompanyreceivedaninjectionof$70millionfromTwitterin201657andinfall2016theFinancialTimesreportedthatSpotifywasinadvancedtalkstopurchasethecompany.
58Thesearchforasustainablebusinessmodelinthissectorcontinues.
Eventhelargestcompaniescontinuetostrugglewithprofitability.
Spotify,withits30millionpayingusersin2015representingabout30percentofitsuserbase,reportedlossesofalmost$200millionin201559,Pandora'slosseswerereportedatover$170million60,Rhapsody'sat$35million61,andSoundCloudisreportedtohavespent$63.
8milliontogenerate$19.
7million.
62Profitabilityeludesallplayersinthissector,forthesereasons:Licensingcostsremainhigh,withindustryleadersPandoraandSpotifyreportingasmuchas70percentofrevenuesbeingpaidouttorightsholders63,whoaregeneral-lythemusiclabels.
Mostusersoptforthefree,ad-supportedserviceasopposedtothead-free,freemi-umservice.
HighsalesandmarketingcostsasnewentrantssuchasApple,Amazon,andTidalenterthespace.
6.
1.
5GamesThevideogameindustry,oncedominatedbyconsole-based,offlineplaycontinuestoshifttoonlineplayandrevenue.
Forexample,Activision's201510-Kfilingreportedanincreaseof$605millionforrevenuesfromdigitalonlinechannels,representing57per-centofthecompany'stotalrevenues.
Thisfigurecomparesto43percentinthepreviousyear.
Similarly,EA(ElectronicArts)reporteddigitalrevenueof$2.
4billionin2015,or57percentoftotalrevenues.
57http://www.
nytimes.
com/2016/06/15/business/media/twitter-invests-70-million-in-soundcloud-music-service.
html58https://www.
ft.
com/content/d03bedbe-85bb-11e6-8897-2359a58ac7a559http://www.
musicbusinessworldwide.
com/spotify-revenues-topped-2bn-last-year-as-losses-hit-194m/60http://www.
musicbusinessworldwide.
com/pandora-losses-hit-170m-last-year-as-listeners-shrunk/61http://www.
billboard.
com/articles/business/6897287/rhapsody-losses-double-2015,March2,201662http://www.
billboard.
com/articles/business/6873893/soundcloud-earnings-2014-19-7-million-burning-cash-building-future63http://www.
musicbusinessworldwide.
com/spotify-contract-three-major-labels-wants-pay-less/66Inthesectorofgamesplayedonsocialplatforms,thepreviousdarlingoftheindustry,Zynga,experiencedarevenuedropofapproximately33percentbetweenour2012studyandthisreport,andKingDigital,thecompanybehindapproximately180titles,themostpopularofwhichisCandyCrush,wasacquiredbyActivisionBlizzardin2015.
NotethatweaccountforthegamingrevenuesofMicrosoftandSonyintheirrespectivecategoriesinthisreport.
OurestimateforGamesis$3.
7billionand8,683jobs.
Table6.
1.
5:Games6.
1.
6OnlineVideoThemostsignificantshiftbetweenthe2012studyandthisreportisthemainstreamingofOTTTelevision,whichstandsfor"Over-the-Top,"referringtothedeliverymechanism,whichisnotthetraditionalbroadcasttelevisionsignalbuttheopeninternet,suppliedbyahousehold'sormobiledevice's(smartphoneortablet)internetserviceprovider.
Thereareover100OTTservicesoperatingintheU.
S.
,mostofwhicharead-freeandsub-scription-based,whilesomeareadvertisingsupported,suchasYouTube.
Thetopcom-paniesinthissectorintheU.
S.
areNetflix,Hulu,Amazon'sPrimeVideo,Facebook'srecentlylaunchedvideoplatform,andGoogle'sYouTube.
Netflixaloneiscurrentlyre-sponsibleformorethan35percentofpeakinternettrafficintheU.
S.
64andonlinevideoasacategoryisprojectedtomakeupabout70percentofmobiletrafficand80percentoffixeddatatrafficby2018.
65NotethatoursegmentcalculationbelowincludestherevenuesandemploymentthatcanbeattributedtothecreatorswhosevideospopulateYouTube.
ActivisionBlizzard$1,1861,856ElectronicArts$9781,881Zynga$5061,360Take-TwoInteractive$366804Ubisoft$3582,400NintendoofAmerica$2793822016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees64http://www.
statista.
com/chart/1620/top-10-traffic-hogs/65http://digiday.
com/platforms/ott-video-going-5-charts/67OurestimateforOnlineVideois$6.
35billionand38,796jobs.
Table6.
1.
6:OnlineVideoIn2015YouTube,withitsoveronebillionusers,reachedmore18to34yearoldsand18to49yearoldsthananyU.
S.
cablenetwork.
Theonlinevideoplatform'sU.
S.
reve-nuehasjumpedfromourestimateof$640millioninthe2012reportto$4-to-$5billionin2015.
Thenetworkofindividualvideocreators,orYouTubers,hassimilarlyexpandedataremarkablerate.
Thesecreators—alsoreferredtoas'YouTubers'—workacrossgenressuchascomedysketches,candidcamerapranks,beautytips,foodhow-to's,'fascinatingfacts'lists,andvideogamecommentary.
AsourcecontactedatYouTubereportedthatmillionsarenowparticipatinginYouTube'sPartnerprogram,inwhichadrevenuesaresplitbetweencreatorsandtheplatformattherateof55percentand45percentrespectively,andthatthenumberofYouTubechannelsearningsixfigureshasincreasedby50percentyearoveryear.
Tables6.
1.
6aand6.
1.
6bbreakouttheearnings,genres,andcountriesoforiginoftheyear'sYouTubers.
Table6.
1.
6a:Top10YouTubePartners2016Netflix$3,0512,231Hulu$1,5001,000Vevo$210375YouTubepartnersandcontributors$1,58435,1902016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployeesTopYouTubersPewDiePie(Sweden)Genre:GameCommentary$15.
0RomanAtwood(U.
S.
)Genre:Comedy$8.
0LillySingh(Canada)Genre:Comedy$7.
5Smosh(U.
S.
)Genre:Comedy$7.
0TylerOakley(U.
S.
)Genre:Entertainment$4.
5RosannaPansino(U.
S.
)Genre:Food$6.
0Revenue($M)68Source:https://yourstory.
com/2016/12/forbes-2016-youtube-celebrities/Table6.
1.
6b:Top100YouTubeChannelsByCountrySource:http://www.
tubefilter.
com/2016/09/19/top-100-most-subscribed-youtube-channels-worldwide-august-2016/Markiplier(U.
S.
)Genre:GameCommentary/Comedy$5.
5GermánGarmendia(Chile)Genre:Comedy/Music$5.
5RhettandLink(U.
S.
)Genre:Comedy$5.
0ColleenBallingera.
k.
a.
MirandaSings(U.
S.
)Genre:Comedy/Music$5.
0UnitedStates46Brazil11GreatBritain7Canada5India4Mexico4Spain4Netherlands3Australia2Chile2Norway2Ukraine2Argentina1Colombia1ElSalvador1Germany1Ireland1Jamaica1Russia1Sweden12016Top100YouTubechannelsbycountryoforigin#ofchannels69WehavecalculatedtheU.
S.
revenueforYouTubersfor2016as$1.
58billion,andhaveexcludedthepercentagespaidtoMCNs(multi-channelnetworks)suchasMaker,Studio71,Machinima,orStyleHaul.
Tomeasuretheimpactofthisnewareaofendeav-orontheU.
S.
economyweestimatedrevenuebasedonthenumberofFTE(fulltimeequivalent)employeescontributingtothesector.
Ourrevenueestimateof$1.
58billionassumeseachFTEemployeeearns$48,000,themedianincomeforaU.
S.
workerin2015.
WhenexpressedasFTEworkers,YouTubecreatorsrepresent35,190U.
S.
jobsin2015,amomentousjumpfromtheapproximately1000FTEsattributedtoU.
S.
YouTu-bersinourpreviousstudyconductedfouryearsago.
6.
1.
7E-learning/OnlineEducationSincethedaysofthepopularizationoftheconsumerinternet,onlinelearninghascarriedthepromiseofademocratizingofeducationandtraining,breakingaccesstoknowledgeawayfromthepreviouslyexclusivedomainofinstitutionsandcampuses.
Anyonewhocouldgetontheinternet,whetherathome,work,orinpublicfacilitiessuchaslibrariesandsocialagencies,couldalsohavetheabilitytotakecourses,class-es,ordiplomaprograms.
OurestimateforE-learning/OnlineEducationis$6.
7billionand58,387jobs,withclosetohalfoftherevenuesascribedtothetoptwoproviders.
Table6.
1.
7:E-learning/OnlineEducationApolloEducationGroup$1,93725,704DeVryEducationGroup$7655,343CareerEducationGroup$6875,416GrandCanyonEducation$6232,555BridgepointEducation$5566,890ScholasticCorporation$5012,535CapellaEducationCompany$4302,887StrayerEducation$308366AmericanPublicEducation,Inc.
(APEI)$2972,866HealthStream$209972Skillsoft$2004672016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees70Inour2012reportinadditiontotheassortmentofonlinecoursesandclassesofferedbysuchcompaniesasBridgepoint,Capella,andRenaissance,wenotedthearrivalofe-learningproviderssuchasKhanAcademy,whichoffersitsservicesforfree,supportedbygrantsfromfoundationssuchastheGatesFoundation.
InfactBillGateshimselfandhischildrenareavidusersoftheservicethatpromotesaccessible,lifelonglearning.
66KhanAcademynowboastsclosetotwomillionsubscribersonYouTubeandisnearingonebillionviewsonitschannelwithvideosontopicsrangingfromphysicstocalculus,history,andeconomics.
OthernewentrantsinthissectoraretheMOOCs(MassiveOpenOnlineCourses),usuallyhostedbyuniversities.
MOOCsofferhighqualitycoursesonline,andgenerallydosowithanopenenrolmentpolicy,i.
e.
prerequisitesarenotrequired.
Feesaregener-allynotchargedtoparticipatethoughforanadditionalpaymentofapproximately$50acertificateofcompletionismadeavailabletostudents.
Amongthemostwell-knownMOOCprovidersareedXandCoursera,withcoursesandinstructorshailingfromtopinstitutionssuchasHarvard,MIT,andStanford.
Intheyearsbetweenthisreportandthelast,firmssuchasUdemyandUdacityhaveemerged,theformerofferingaplatformonwhichanyonecancreateandteachacourse,andthelatteroffering'nanodegrees'(certificationforspecificwork-relatedskills).
However,theseandothere-learningfirmsexaminedarenotyetlargeenoughtowarrantinclusioninthisstudy.
E-learninghasexperiencedperiodsoftremendousgrowthoverthepast15to20yearsofitsexistencebuthasalsoexperiencedpullbackinsomeareas.
Despitethesubstantialsectorrevenuesandemploymentnumbersreportedabove,thefollowingtwoexamplesprovideillustrativeifnotcautionarytaleswhenitcomestoonlineeducation,particularlywhenaccesstofederalfundsisinvolved.
ITT,anationallyaccreditedinstitutionwithafocusonbusinessortechnicalcourses,hadcloseto$1billioninrevenuein2015.
Nonetheless,duetoquestionablebusi-nesspractices,theU.
S.
DepartmentofEducationbannedtheschoolfromacceptingstudentsusinggovernment-providedfinancialaidinthesummerof201667andbythefallof2016ITThadcloseditsdoors.
66http://www.
inc.
com/lisa-calhoun/bill-gates-uses-this-tool-to-teach-his-kids-should-you.
html67http://www.
ed.
gov/news/press-releases/department-education-bans-itt-enrolling-new-title-iv-students-adds-tough-new-financial-oversightRenaissanceLearning$162670ITTEducationalServices$1461,441Follett$13527571ApolloEducationGroup,betterknownastheparentcompanyofUniversityofPhoe-nix.
AtonetimethecompanywasamongGoogle'slargestclientsforsearch-basedadvertising,explainingitsubiquitouspresenceonwebsites,blogs,andYouTube.
Thesizeofitsstudentbodydecreasedfrom460,000to213,000in2015,andgradua-tionrateswerereportedasbeinginthesingledigits.
686.
2E-commerceWereportone-commerceinthefollowingsegments:Onlineretailing(combiningpureplaydigitalretailingandhybridonline/offline)OnlinetravelservicesOnlinefinancialservicesincludingday-to-banking,investment,digitalpaymentsandcurrencies,andthenewsectorofFinTech,orfinancialtechnology6.
2.
1RetailingMostofthenearly$5trillioninU.
S.
retailsalesoccursinstores.
In2016,retaile-com-merceaccountedforjust7.
1percentofallretailsales.
69Howeveritwaswhereover60percentofretailgrowthtookplace.
ThetradepublicationinternetRetailerreportsthate-commercegrewby15percentintheU.
S.
in2015,whileretailsalesgrewbyjust1.
5percent.
70Inthetablethatfollowsweidentifytheonlinerevenuesoffirmsinthetop10ofonlinee-commerce,androllupseparatelythosefrom11to500andthosebeyondthe500th.
Wethengiveourestimatesoftheaggregateretailsalesbyindividualsandsmallfirmssellingone-commerceplatforms.
Ourestimatefore-commerceactivitybyretailersis$319.
8billionand770,211jobs.
68http://money.
cnn.
com/2015/03/25/investing/university-of-phoenix-apollo-earnings-tank/69http://www.
emarketer.
com/Article/U.
S.
-Retail-Sales-Near-5-Trillion-2016/1013368#sthash.
pEZAKvxY.
dpuf70https://www.
internetretailer.
com/2016/02/17/us-e-commerce-grows-146-201572Table6.
2.
1:TopU.
S.
E-commercefirmsOurestimatebeginswithenterpriseretailingontheinternetandincludesbothpureplayonlineretailersandofflineretailerswhosellonlineappearinthiscategory.
Also,weassume,basedonasampleofpureplayonlineretailers,thateachemployeegenerates$420,000ofrevenueandmakeourestimatesaccordingly.
Therevenueandemploymentinthiscategorybreaksdownasfollows:ThethreelargestU.
S.
onlineretailersareAmazon,Apple,andDelland,becausetheyhaveotherlinesofbusiness,arenotrecordedhere.
AmazonistreatedinthechapteronIntegratedFirms,andAppleandDellareanalyzedwithintheHardInfra-structurechapter.
ThetotalU.
S.
e-commercerevenuesfortheremainderoftheTop10retailersis$32.
08billionandourinternet-dependentemploymentestimateis76,386.
WeestimatethetotalforU.
S.
e-commerceforretailers11through500in2016as$176.
9billionand421,283jobs,andassumeanadditional$11billionofrevenueand27,500employeesforU.
S.
e-commerceforretailers500andbelow.
AmazonA$57,337122,324AppleB$12,00028,571DellB$8,50020,238Walmart.
com$8,35519,892Macys.
com$5,58113,288Staples.
com$4,73011,262HomeDepot.
com$4,26810,161Costco.
com$3,2907,833OfficeDepot.
com$3,1947,604QVCGroup$2,6656,346Retailers11to500$176,939421,283Retailersbeyond500$11,00035,000AAccountedforinIntegratedFirmschapterBAccountedforinHardInfrastructurechapter2016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees73NextweaccountforsellersonsitessuchasAmazon,eBay,Craigslist,andEtsyintheU.
S.
,inthefollowingtable:Table6.
2.
2:IndividualsellersFulltimeemploymentestimateswerearrivedatfromestimatesofthetotalnumberofsellersperplatform,andapplyingtheU.
S.
SocialSecurityAdministrationestimateofme-dianpersonalincome,lessplatformcommissions(whererelevant)andpaymentprovidercommissions(whererelevant).
Craigslist,eBay,andEtsyhavealwaysbeenhometoindividualmerchants,rangingfromone-offsellersofsaleablegoodsinatticsandgaragestopeoplelikeRyanFinley,who,onthesiterecraigslist.
com,documentshowhemakeshisfull-timeliving,support-ingafamilyofseven,bybuying,repairing,andthenre-sellingitemsonCraigslist.
"It'stheonlybusinessI'veeverstartedthatdidn'tfail,"Finlaywroteinablogpost,71andhissuccessasanindependentonlinemerchantisfarfromisolated.
Inasimilarvein,Amazonhasshiftedfrombeingprimarilyaplatformfortraditionalretailconductedintheonlineenvironmenttoaplatformforsmallonlinesellers,abletoflextheirentrepreneurialmusclebybecomingthirdpartymerchants.
Infact,halfofthegoodssoldatAmazon.
comnoworiginatewiththirdpartymerchants,whoarereportedtonumbertwomillionwithsalesinexcessof$130billion.
72InJeffBezos'annuallettertoshareholderstheAmazonCEOreportedthatofthetwomillionthirdpartymerchants,approximately70,000areearningsalariesofmorethan$100,000peryear.
Bezosalsopointedoutthatthisgrowthofthenumberofthirdpartymerchants,andtheirrevenuenumbers,canbeattributedtoentrepreneursusingtheplat-formtocreatetheirownlinesofbrandedproducts,thusallowingthemtochargehigherpricesthanmerchantssimplyresellingitems.
73Amazonsellers$247,98545,500eBaysellers$34,302187,489Etsysellers$1,68035,339Craigslistsellers$60013,3332016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees71https://priceonomics.
com/post/16529584021/how-to-make-it-on-craigslist72http://www.
seattletimes.
com/business/amazon/amazon-to-host-forum-for-its-marketplace-merchants/73http://www.
seattletimes.
com/business/amazon/amazon-to-host-forum-for-its-marketplace-merchants/74AlsomeritingacloserlookinthissegmentaretheoperationsofCraigslist.
ThenofrillssitebeganlifeasamodestemaillistofthingstodoinSanFranciscointhemid-1990sandhassincereplacedtheclassifiedssectionofnewspapersworldwide,oncea$16billionmarketplace.
74TheimpactofCraigslistontheprintnewspaperindustryhasbeendramatic—revenuesfromclassifiedadshavereportedlydroppedbyover70percentsince2000.
75AnnualrevenuesatCraigslisthavetripledto$343millionsinceourlastanalysisconductedin2012andhavedoubledinthepastyear,allthisachievedwithaleancorestaffof40inthecompany'sSanFranciscoheadquarters.
Craigslist'sgrowthcanbeattributedtothecontinuedmainstreamingoftheonlinesellingsiteaswellasanewfeeof$5beingchargedforautoandtruckadsbydealers,saidtobringinanother$40millioninannualrevenueontopofthefeesalreadybeingchargedforjobpostingsinselectcities,apartmentbrokersinNewYorkCity,therapeuticservicesintheU.
S.
,andticketsalesbydealersintheU.
S.
76Whilemanyofthetoponlineretailersearnedrevenuesinthetensofbillions,ouranal-ysisalsoincludescompanieswhoseplayingfieldcanbefoundinhighlypersonalized,customizedgoodsandservices.
Thisdigitalenablingofphysicalgoodsspecializinginnicheareashasledtothesuc-cessofsuchfirmsas:Blurb,whichenableson-demandanddo-it-yourselfbookpublishingoflongtail,akaniche,topics.
Forexample,thecompanyissaidtoberesponsiblefor"theworld'slargest"collectionofbooksaboutwomen'srollerderbyteams.
"77Blurbisprofitable,withrevenuesofover$100millionsince2014.
78CaféPress,wheretherangeofcustomizablegoodsgoesfarbeyondt-shirts,base-ballcaps,andmugstoincludeshowercurtains,lunchbags,licenseplateframes,andphonecases.
Stamps.
com,acompanythatbringstheconvenienceofonlinepurchasingtooneofthemostinconvenienttasks,i.
e.
buyingstamps.
Thispostageon-demandservicereportedrevenuesof$214millionin2015.
74http://3taps.
com/papers/Craigslist-by-the-Numbers.
pdf75http://www.
forbes.
com/sites/stevenrosenbaum/2015/01/26/the-craigslist-economy-is-booming/76http://www.
craigslist.
org/about/help/posting_fees/77http://fortune.
com/2014/05/06/blurb-acquires-hps-magcloud-aims-to-dominate-long-tail-publishing/78http://fortune.
com/2014/05/06/blurb-acquires-hps-magcloud-aims-to-dominate-long-tail-publishing/756.
2.
2TravelServicesThemainstreamingandaffordabilityoftabletsandsmartphones,combinedwiththegrowthoftheappmarketplacehasledtoanarrayofnewconsumerbehaviorsrelatedtoresearchingandbookingtravel.
Thissegmentexaminesthefirmsthatprovideinternet-dependenttravelservicestoU.
S.
consumers,comprisingtravelcomparisonshopping,travelreviewsites,aggregators,andtravelbookingdoneusingtheinternet,includingbothtraditionalofflinetravelagen-ciesusingtheinternetanddigital-onlyagenciessuchasExpediaandPriceline,referredtoinindustryparlanceasOTAs(OnlineTravelAgents/Agencies).
Wealsoincludeflightsbookeddirectlywithairlinesontheirownsites,asubstantialandgrowingsourceofairlinerevenue.
Whilefromtheconsumer'spointofviewTripAdvisoristhoughtofasareviewandrecommendationsite,weincludeitinthissectionasthebulkofitsrevenuescomefromclick-basedadvertisingwhichdirectuserstothesitesofhotels,airlines,cruisecompa-nies,andOTAs.
79OurestimateforTravelServicesis$55.
1billionand23,943jobs.
Table6.
2.
2:TravelServicesSouthwestAirlines$15,431926AmericanAirlines$8,364502DeltaAirlines$8,262496UnitedContinentalAirlines$6,711403Expedia$3,70310,489AlaskaAirlines$3,023181Priceline$1,8173,200JetBlue$1,53792VirginAmerica$87252TripAdvisor$7011,805CheapOair$2741,400ThomasCook$109218Travelzoo$89312MaritzTravel$641272016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees79http://ir.
tripadvisor.
com/secfiling.
cfmfilingID=1564590-16-1286276Thoughthereareover100airlinesoperatingintheU.
S.
thetop10carried80percentofthepassengersin2015andthetopfour—American,Delta,Southwest,andUnited(nowUnitedContinental)—accountfor60percentofthemarket.
80OuranalysisalsoincludesthreeofthesmallerU.
S.
carriers—Alaska,JetBlue,andVirgin—andaccountsfortheremainderofairlinesinan'allother'category.
InordertoavoidpayingcommissionstoOTAsinahighcompetitive,price-sensitiveindustrymanyairlinesarenowpromotingdirectbookingsontheirownwebsites.
SomeairlinesreporteddirectbookingstotheirsitesintheirannualSEC10-Kfilings,andwewerethereforeabletousethosefigures,suchasthoseofAlaskaAirlines,whichbrokedownitssaleschannelas60percentthroughalaskaair.
com,23percentthroughtradi-tionalagencies,11percentthroughOTAs(Expedia,etal.
),andsixpercentthroughthereservationcallcenter.
SouthwestAirlinesisparticularlynotableinthiscategory,withits10-Kfilingreporting79.
4percentofitspassengerrevenuesoccurringviaitsownweb-site,asSouthwestdoesnotusetheOTAstogenerateleadsandsales.
Forthosecarriersthatdidnotseparateoutbookingsdirecttotheirownwebsiteintheir10-Kfilingsweusedthebenchmarkof34percentprovidedinthe2015study"BenefitsofPreservingConsumers'AbilitytoCompareAirlineFares.
"816.
2.
3FinancialServices:Banking,FinTech,DigitalPaymentsandDigitalCurrenciesInourlastreportpublishedin2012thebigshiftnotedinthebankingindustrywastwo-fold:Theeclipsingoftheopeningoffinancialproductsintheonlinevs.
thebranchenvi-ronmentThetopfivebankingactivities—billpayment,viewingbalances,viewingstatements,retrievingtransactionhistories,andtransferringfunds—tookplaceprimarilyonline.
Intheyearssince2012,thebankingindustryhasundergoneadditionalsignificantshiftsastheproliferationofsmartphoneshasenabledfunctionssuchasthephotographingofchecksfordeposits,thusfurtherreducingtheneedforindividualstovisitbanksinper-son.
OurestimateforFinancialServicesis$26.
3billionand72,233jobs.
80viaMintelJuly2016ReportonU.
S.
AirlineIndustry81Morton,F.
,"BenefitsofPreservingConsumers'AbilitytoCompareAirlineFares,"May2015,http://www.
traveltech.
org/wp-content/uploads/2015/05/CRA.
TravelTech.
Study_.
pdf,p.
8.
77Table6.
2.
3:FinancialServices:Banking,FinTech,DigitalPaymentsandDigitalCurrenciesToarriveatthesefiguresweusedthefollowingmethod:First,weperformedtwoindustry'rollups',oneforthetopfourbanksintheU.
S.
in2015,andasecondfortheremain-derofU.
S.
banks.
Thelogicforproceedinginthisfashionisthatthefourlargestfirmsclaim27.
2percentofthemarket,withthemarketshareofthefifthfirmbeinglessthanonepercent.
82Wethereforeperformedananalysisoftheremaining72.
8percentofthemarketwithan'allother'category.
ThetopfourbanksanalyzedwereWellsFargo,BankofAmerica,JPMorganChase,andCitibank.
Thetotaloftheir2015globalrevenuesis$145.
3billion,fromwhichwededuct$40.
83billionasrevenueduetocreditcards,accountedforseparatelyinthisstudy.
Oftheremaining$104.
5billionweestimate$90.
9billion,or87percent,asU.
S.
rev-enue.
Thefigureof87percentisbasedonareviewoftherespectivecompany's10-KfilingswiththeSEC(SecuritiesandExchangeCommission).
The10-Kfilingsalsopro-videduswiththebasisforourestimateof$380,000ofannualrevenuegenerationperemployee.
Ourestimatefortheinternet-dependentU.
S.
revenueforthetopfourbanksis$891millionandweascribe1,273FTE(fulltimeequivalent)employeestothisrevenuefigure.
Weperformedasimilaranalysisforthe'allother'categoryofU.
S.
banksandourfind-ingswere$2.
99billionininternet-dependentU.
S.
revenueand4,279FTEjobs.
Banks1to4$5091,273Banks(AllOther)$2,9954,279CreditCardsRollUp$11,80829,520FinTechRollUp$3,37517,707FinancialServicesRollUp$4,67614,017E-Trade$1,4283,400InteractiveBrokersGroup$1,097473CharlesSchwab$181542Blucora$118772TDAmeritrade$1052502016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees82IBISWorldU.
S.
IndustryReports-CommercialBanking201578Ourcalculationforinternet-dependentemploymentinthissectorusesthesumofpeopleemployedinthedigitalmarketingdepartments(i.
e.
thoseinvolvedinonlineadvertising,socialmediamarketing,mobilemarketing,emailandonlinechatcustomerservice),aswellasthosewhoworkinonlinebankingoperationsandonthemobileappteams.
Foremployeesindigitalmarketingweassume100employeesper$50billionofrevenue,basedonemploymentfiguresforsimilarlyscaledretailers.
Forthoseinvolvedinonlinebankingandmobileappoperationsweassume600peopleper$50billionofreve-nue.
83Theevolutionofworkflowinthebankingindustryfromface-to-faceencounterswithtell-ersinbrickandmortarbankofficestoonlineandmobiletransactionstimesandplacesconvenienttocustomershasmerelybeenphaseone.
Inthepasthandfulofyears,wehaveseentheemergenceoftheFinTech(shortforfinancialtechnology)sector,amultibil-liondollarmarketattractingvigorousventurecapitalfundingandexperiencingexponen-tialgrowthinconsumeruptakeandrevenues.
TakeforexamplethecreditmarketplaceLendingClub,whichin2015managed$460billioninassets,employed1000people,andhadrevenuesof$730millionwithabusinessbasedontechnology-enabledpeer-to-peerlendingthatoffersborrowersbetterratesthantraditionalbanksandprovideslenderswithmoreattractivereturnsthanthefractionofafractionofapercentagethathasbeenthenormforbankdepositsoverthepastseveralyears.
Asevidencedbythisbusinessmodeltheimpactoftheinternetonthebankingindustrynowgoesfarbeyondthemeredigitizationofconventionalbankingfunctionsandshiftstowhatcouldbethoughtofasradicalalternatives,ifnotsystemicthreats,tothelegacyprocessesthatdefinedtheindustryfordecades.
Thestatusofthephysicalbankthatoc-cupiedabuildingresemblingaGreco-Romantempleandoperatedinaformal,almostgovernmental,manner,isnowunquestionablybeingchallengedbycompaniesthatoperateentirelyinthedigitalsphereandofferefficienciesunimaginableinapre-internetworld.
FinTechcoversbusinessesrangingfrom'robo-advisors'thatofferautomatedpersonalwealthmanagementforafractionofthepriceoffinancialadvisors(usually0.
25per-centinlieuofthemutualfundindustryaverageof1.
3to1.
5percent),peer-to-peerlendingpoweredbypredictiveanalyticsandalgorithms,digitalpaymentsystemsthatallowindividualpayments,grouppayments,andevenmicropayments,andcrowdfund-ingplatformsthatsupporteverythingfromartisticendeavors(e.
g.
Kickstarter,Indiegogo)tohoneymoons(Honeyfund)topersonalneeds,charities,andcommunityprojects.
Inthecaseofthelattersegment,GoFundMe,theprimaryplatformofitskind,nowraises$100millionpermonthforsuchnon-commercialactivities,withthecompanyretainingafivepercentfeeforfacilitatingtransactions.
83Forresterreport"StaffingandHiringforeBusiness,"201179Inthissectorwealsofindinnovativebusinessmodelssuchasthefollowing:CircleUp,whichenablesinvestinginconsumerproductstartups.
Coinbase,theworld'slargestdigitalcurrencycompany,usedbyconsumers,mer-chants,andsoftwaredevelopers,with3.
8millionconsumerwalletsandover$3.
5billioninBitcointransactions.
CommonBond,amarketplaceforstudentloansthatoffersbetterratestoborrowersandbetterreturnstolenders.
Kabbage,offeringfullyautomatedfinancialservicestosmallbusinesses.
Mozido,amobilepaymentplatformfocusingonthetwobillionownersofmobilephoneswhoare'unbanked'(i.
e.
donothavebankaccounts).
Inadditiontotheseswathsofinnovationinthefinancialindustry,anothersizeablechangeonthehorizonistheblockchain,a.
k.
a.
thedecentralizedledgerbehindsuchdigitalcurrenciesasBitcoin.
Itisestimatedthatby2017itwillbeusedby15percentofbanksforactivitiesrangingfrompaymentsandlendingtoreal-timeinformationsharingwithintheenterprise.
"846.
3TheOn-DemandEconomyAnewworldofflexibleworkhasemergedbetweenthetimeofourpriorreportandthisone.
Itisdistinctfromtheworldoffreelanceorcontractorworkforitsabilitytobecon-ductedatthediscretionoftheworker,in,e.
g.
,thesparefewhoursonaThursdaynightoraSundayevening,orcanbepursuedfulltime,withtheindividualworkerabletosethis/herhours.
Sowidespreadhason-demandeconomyworkbecomethatin2016theU.
S.
CommerceDepartmentproposedanewclassification,'digitalmatchingfirms',inanattempttocreateamoreaccuratedescriptionoftheworldofplatform-enabledflexiblelabor.
85Theeconomicfoundationoftheon-demandeconomyistheplatformbusiness,which,ratherthanbuildphysicalassetsandinfrastructure,buildsanonlinedestinationthroughwhichbuyersandsellerscantransact.
PlatformRevolution(2016)authorsParker,VanAlstyne,andChoudarydescribeplatformsas"…bring[ing]togetherproducersandconsumersinhigh-valueexchanges.
Theirchiefassetsareinformationandinteractions,whichtogetherarealsothesourceofthevaluetheycreateandtheircompetitiveadvantage.
"8684http://fortune.
com/2016/09/28/blockchain-banks-2017/85http://fortune.
com/2016/09/15/gig-freelance-economy-size/86https://hbr.
org/2016/04/pipelines-platforms-and-the-new-rules-of-strategy80Characteristicsoftheplatform-enabledon-demandeconomyinclude:Accessoverownership(consumerpointofview)Flexibilityoverfixedhours(workerpointofview)Optionsandmodularityofchoice–e.
g.
aspareroominsteadofhotel,asharedrideinsteadofacabWedividethiscategoryintotwosections:Theplatformsthatenableon-demandgoodsandservices(andtherevenuesthesefirmsderivefromfeesandcommissions)Theworkerswhoprovidetheon-demandservicesthatpowerthisneweconomy6.
3.
1On-DemandPlatforms&ServicesThefirmsthatmakesuchworkpossibleareknownasplatformfirms,which,ratherthanbuildtheinfrastructureoftaxifleetsordeliveryservices,createaplatformformatchingbuyersandsellersofservices.
Amongthemostwell-knownofsuchfirmsareAirbnb,taxialternativesUberandLyft,anddeliveryon-demandservicesInstacart,Grubhub,Task-Rabbit,andPostmates.
OurestimateforPlatforms&Servicesis$2.
4billionand12,803jobs.
NotethattheindividualswhoprovidethegoodsandservicestotheplatformcompaniesaredetailedinSection6.
3.
2,entitled"On-DemandEconomyWorkers.
"Table6.
3.
1:Platforms&ServicesUberTechnlogies$8656,700Lyft$5253,000Airbnb$4501,184Grubhub$3581,094Postmates$100350Instacart$100350TaskRabbit$41252016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees816.
3.
2On-DemandEconomyWorkersInadditiontodriversfor,e.
g.
,UberandLyftanddeliverypersonsfor,e.
g.
,TaskRabbitandPostmates,wealsoincludepeopleintheU.
S.
whoserveasAirbnbhostsinourcalculation.
NotethatwedonotincludethemerchantsandartisanssellingonplatformssuchaseBay,Etsy,Amazon,andCraigslistinthiscategory,butwedoaccountfortheminthee-commercesectionofthisreport.
OurestimateforOn-DemandEconomyWorkis$6.
04billionand134,160jobs.
Overall,ourresearchsuggeststhatthereare54millionto68millionpeoplewhoareindependentearners,tovaryingdegrees,andwithvaryingdependenceontheinternet,intheUnitedStates.
ThisfiguremaybecomparedwiththatofMcKinseyGlobalInstitute,whose2016reportentitled"IndependentWork:Choice,Necessity,andTheGigEcon-omy"foundthatthissegmentofworkersissignificantlylargerthanpreviouslythought.
McKinseyestimatesthatupto30percentofworkersinWesternEuropeandtheUnit-edStates(162millionpeople,ofwhichalmost60millionarelocatedintheU.
S.
)areindependentworkers,coveringthegamutfromthosewhosupplementtheirmainincomewithahandfulofhoursofon-demandeconomyworkweeklytothosewhoareabletoperformsuchworkasaprimarysourceofincome.
87TheMcKinseyreportdivideson-demandeconomyworkersintofoursegments:30per-centwhoarefreeagentsbychoice,40percentwhoarederivingsupplementalincomefromtheiractivities,14percentwhomaketheirprimarylivingwithon-demandwork(butwhowouldprefertraditionalemployment),and16percentwhoareinfinanciallyunstablesituationsandperformtheworkoutofeconomicnecessity.
Toarriveatourestimate,weusedthefollowingmedianfigures:Airbnbhosts:$4,100perannum88On-DemandDrivingandDeliveryServices:$10-15/hourly89($30,000perannumFTE).
WethenascribedadiscountedSSA(SocialSecurityAdministration)medianincomeof$30,000tothefirstcategoryof30percentfreeagents.
Wethenassumedthatthe40percentwhoaresupplementalincomeworkersdosoforanaverageofeighthoursperweekat$15/houror$480permonthx12months,foranannualincomeof$5,760,or11.
7percentoftheFTEsalaryof$48,880.
Theremaining30percent(thesumofthe14percentand16percentinthethirdandfourthcategories)havebeenattributedadiscountedSSAmedianincomeof$15/hour.
87http://www.
mckinsey.
com/global-themes/employment-and-growth/independent-work-choice-necessity-and-the-gig-economy88http://bgr.
com/2016/07/08/airbnb-profits-new-york-city-annual-revenue/89http://uberdriverdiaries.
com/how-much-do-uber-drivers-really-make/82Toassistuswithdistributingtheon-demandeconomyworkgeographicallywefactoredinthefollowingfindings:"Between2010and2014,the25largestmetrosaccountedformorethan80percentofthenetgrowthingigeconomyfirmsintheride-sharingsector,with90percentconcentratedinthelargest50metros.
"90"In2015…atotalof550,000AirbnblistingsintheUnitesStates[wereidentified].
Californialeadsallstateswith125,803totalpropertieslisted,withNewYorkinsecondplaceat94,976Airbnbrentals.
"916.
4SocialNetworks&ServicesThiscategoryincludesthesocialplatformsineverydayusebyhundredsofmillionsintheU.
S.
,themostpopularofwhichprovideforumsforentertainment,artisticexpression,businessnetworking,andofcoursesocializing,whetherondesktop,tablet,orphone.
Ourdefinitionofasocialplatformisoneinwhichtheuser'sexperienceisimprovedbythenumberandqualityofconnectionss/hehas.
Wealsoincludeasasubcategorytheonlinedatingsector,whichinitiallymademeetingmatesaseasyasaclickonaweb-site,andmorerecentlyhasintroducedfeaturesoptimizedformobile,suchastouch-screenswiping,locationfeatures,andintegrationwithone'ssocialnetworks.
6.
4.
1SocialMediaSitesWhatbeganaspastimesindormroomsorwildideashatchedbySiliconValleystart-ups,socialmediasiteshavesincebecomethedefactohomepagesoftheinternetforbillionsofpeopleworldwide.
PlatformssuchasFacebook,Twitter,andSnapchatarethe'goto'onlinedestinationforthosewishingtosocialize,discussanddebate,orjustpostpicturesandvideos.
Thesesocialsiteshavereplacedtheportalsofthefirstgenerationconsumerinternetandthesearchenginesthatfollowed.
Notethattherevenuesofthelargestsocialplatform,Facebook,areaccountedforinoursectiononIntegratedFirms,towhichFacebookbelongsduetoitsbusinessoperationsoutsideofsocialnetworking.
OurestimateforSocialMediaSitesis$3.
4billioninrevenueand8,893jobs.
90http://www.
citylab.
com/work/2016/10/the-gig-economy-and-the-rise-of-cities-as-platforms/503795/91http://blog.
airdna.
co/2015-in-review-airbnb-data-for-the-usa/83Table6.
4.
1:SocialMediaSitesTwitter,thoughoftencriticizedforitsinabilitytoturnaprofit,alsogrewsignificantlyintheyearsbetweenourtwostudies.
In2011thecompany'sU.
S.
revenuewasestimatedtobeamere$20million,whereasby2015thatfigurehadswelledto$1.
44billion.
TheproblemwithTwitter,then,isnotitsabilitytogeneraterevenue,butitsabilitytocontinuetogrowitsuserbase,whilereininginwhatarereportedlyveryhighsalesandmarketingcosts.
Thesecostsareestimatedatapproximately40percentofrevenue,aratethatismorethandoublethatofcompetitorssuchasFacebookandGoogle.
Similar-ly,thecompany'sR&Dcostsindexhigherthancompetitors,runningatapproximately28percentofrevenues.
92Snapchat,which,atthetimeofourlastreportwasanicheappspecializinginrapidlyvanishingphotoshassincebecomeamediapowerhouse.
ItnowenjoyspartnershipswithmajortraditionalmediapublisherssuchasNBCandNationalGeographicaswellasnewerpublishersViceandBuzzFeedandhassurgedtoover10billionviewsperday.
ThoughthefigureswereportforU.
S.
revenuesandemploymentfor2015areamodest$39millionand218employees,wenoteherethateMarketer'sprojectionsforSnapchat'sadrevenuefor2016and2017are$367and$935.
5millionrespective-ly.
93AnothersocialsitethathasrisentoprominencesinceourlaststudyisPinterest.
Itsnum-bersfor2011were$130,000ofU.
S.
revenueand21employees,whereasfor2015weestimaterevenuesof$88millionandastaffof330.
Thevisualscrapbookingsitecontinuestorampupitsofferingsonthebackendwithavarietyofadtechcompaniesandonthefrontendpairingsocialmediainfluencerswithbrandsandagencies.
94ThoughacquiredbyMicrosoftinthesummerof2016LinkedInappearsasaseparateentityinouranalysisofthestateoftheinterneteconomythroughtotheendof2015.
LinkedInCorporation$1,8465,811Twitter$1,4432,534Pinterest$88330Snapchat$392182016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees92http://www.
reuters.
com/article/us-twitter-m-a-restructure-analysis-idUSKCN12L0EQ93http://adage.
com/article/digital/snapchat-ad-sales-reach-935-million-year-emarketer/305722/94http://www.
adweek.
com/news/technology/pinterest-enlists-top-creators-and-production-shops-craft-posts-marketers-17412184ItgrewfromU.
S.
revenuesof$522millionin2011to$1.
85billionin2015,withtheU.
S.
employeeheadcountrisingaccordinglyfrom1,200to5,811.
Itisalsoworthmentioningthefirmsinthiscategorywhosestarsfellintheyearsbe-tweenourstudies.
News'up-voting'siteDiggwasoneoftheearliestsocialmediaphenomenayetbymid-2012thebrandandtechnology(thoughnotalltheassets)weresoldtoaventurecapitalfirmforjust$500,000.
95Check-insiteFoursquarehadsimilarlybeentoutedasa'nextbigthing'companyatthetimeofourpreviousstudy.
For2015Hooversreportedafairlymodest$50millioninrevenueforthecompany.
6.
4.
2OnlineDatingThedatingmarketplacecomprisesonlineandofflineactivities,withonline(websitesandapps)accountingfor$1.
8billion,or75percentofthe$2.
4billioninU.
S.
revenuein2015,andofflineactivitiessuchassinglesevents,speeddating,andmatchmakersmakinguptheother25percent.
96OurestimateforOnlineDatingis$1.
1billionand2,018jobs.
Table6.
4.
2:OnlineDatingThedatingfirmsthatareindividuallyenumeratedcomprise$458millionofIBIS'ses-timateofindustryvalueof$1.
8billion97.
Afurther$633millioncomes,accordingtoIBIS,fromthedatingsitesofIAC(Match.
com,PlentyofFish,Tinder,OKCupid,andoth-ers.
)Thereforethe"allother"entryintheOnlineDatingtableis$709million.
IBISgivesthisindustry'semploymentas8,296,ofwhichwetake75percenttobeonline,andwecomputethe"allother"employmenttobe1,560afterallowingforIAC'semployment.
eHarmony$312190Zoosk$10074SparksNetworks(jdate.
com,christianmingle.
com,blacksingles.
com,etc.
)$46194Allother$7091,560Note:IACdatingappssuchasMatch.
com,PlentyofFish,andTinder—withcollectiverevenuesof$633M—areaccountedforintheMulti-GenreContentsectionofthischapter.
2016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees95http://www.
forbes.
com/sites/jeffbercovici/2012/07/12/digg-once-worth-164-million-sold-to-betaworks-for-500k/#722978b42c6696viaeMarketer,U.
S.
DatingServiceRevenueSharebyType,June201597IBISWorldReports,U.
S.
DatingServicesApril2016.
85Thestoryofhowthissectorhasdevelopedistwo-fold:Ontheonehand,itisastoryofindustryconsolidationoftheoriginalweb-baseddatingsitessuchasMatch.
com,OKCupid,andPlentyofFish,allofwhicharenowownedbyIAC(InteractiveCorp.
),whichalsoholdsamajoritystakeinTinder.
Ontheotherhanditisthestoryofashifttoapp-baseddatingtailoredtothemobileuserexperienceonsmartphones.
NotethattherevenuesandemploymentofIACarecapturedintheMulti-GenreContentcategoryearlierinthischapter.
Somedatingsitesaresubscription-based,suchasMatch.
comandeHarmony,chargingusersbetween$30and$60permonth.
FreeoptionssuchasOKCupidandPlentyofFishobtainthebulkoftheirrevenuesthroughadvertising,thoughthesesitesoftenofferpremium,ad-freeversionsoftheirservice.
Manyfreesitesalsoofferuserstheabilityto'boost'theirprofilewithadditionalfeatures,ortoseeiftheirmessagetoauserhasbeenviewed,forafee.
Bothdemandandrevenueshavegrowninthissectorsinceourlaststudyasthestigmaofusingdatingsiteshasabatedandniche,ethnic,andreli-gion-baseddatingsiteshaveemerged.
98Ouranalysisofthissectorhastakenthe'topdown'approach,breakingouttheinter-net-dependentrevenuesfortheindustryasawhole,aswellasa'bottomup'approach,examiningtherevenuesandemploymentofthelargerindividualfirms.
Ourestimateforapp-basedmobiledating,thebiggestdevelopmentinthissectorsincethetimeofourlaststudy,is$468million.
99Thesematchingserviceshavebeenopti-mizedforthemobileexperienceofatouchscreen,socialnetworkconnections,loca-tioncheck-insandinterstitialmoments.
Somehavelikenedtheseappstoreallifevideogames,whereplayersarepushedstreamsofphotosandswipelefttoindicatelackofinterestandrighttosignalpotentialinterestandfollowupwithinstantmessages.
AmongthemostpopularofsuchappsareTinder,Bumble,Hinge,andHappn,whichincorporatefeaturessuchasnotifyingyouwhenyouphysicallycrosspathswithsome-oneinreallife,introducingyoutopeoplewithwhomyouhaveFacebookfriendsincommon,orscanningyouriTunescollectionandmakingsuggestionsbasedonsharedmusicaltastes.
6.
5EmploymentandRecruitingThiscategorycomprisesonlineemploymentsites(e.
g.
Indeed,Monster)andthesectorofsoftwarefocusedonhumanresources,payrollandbenefits,andrecruiting.
98IBISWorld,U.
S.
IndustryReports,DatingServices,April201699IBISWorld,U.
S.
IndustryReports,DatingServices,April2016866.
5.
1EmploymentServicesJobhuntingwasoneoftheearliestsectorstomigratetotheinternet,ledbyMonsterinthemid-1990s.
Sincethen,onlinerecruitinghasgrowntoamultibilliondollarbusinessintheU.
S.
,growingannuallybyapproximately15percent.
100NotethatLinkedInisnotincludedinthissegmentasitappearsinthecategoryofsocialnetworksinourreport.
The'allother'categorynotedinthetablebelowincludesfirmssuchasSnagajob,TheLadders,andDice.
OurestimateforEmploymentServicesis$1.
5billionand8,567jobs.
Table6.
5.
1:EmploymentServicesMostjobsitesarefreetouseforjobseekers,whetherpostingorbrowsing,withrevenuestypicallygeneratedthroughthesaleofadvertising,subscriptionswithpremiumfeatures,andfeeschargedtorecruitersandfirmsseekingemployees.
NewtoourreportinthissectorisCare.
com,aplatformformatchingcaregivingser-vicesproviders,frompetcareandhousekeepingtochildcareandseniorcare,withthoseinneedofsuchservices.
NotableacquisitionsinthissectorincludetheNether-lands-basedmultinationalhumanresourceconsultingfirmRandstadwhichacquiredMonster,andMicrosoftwhichacquiredLinkedIn,bothin2016.
6.
5.
2HumanResourcesSoftwaresystemsdesignedforhumanresourcesfunctionssuchaspayrollandbenefits,recruiting,andemployeewellnessprograms,appearinthiscategory.
OurestimateforHumanResourcesis$1.
1billionand5,628jobs.
Monster$4262,368Indeed$3001,200CareerBuilder$1451,500Care.
com$139799Allother$5002,7002016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees100IBISWorldOnlineRecruitmentSitesintheU.
S.
:MarketResearchReport,February201687Table6.
5.
2:HumanResourcesHumanResourcessoftwareisbeingcoveredinourreportforthefirsttime.
InthepastweaccountedforHRandpayrollsoftwareintheNAICScodeof511210/SoftwarePublishing,andexaminedfirmssuchasSabaandSage.
Additionally,weaccountedforHRsoftwarefunctionsintherevenuesoflargefirmssuchasOracleandSAP,whichacquiredcompaniesspecializinginthisspace.
OracleacquiredPeopleSoftandTaleo,andSAPacquiredSuccessFactorsinlate2011for$3.
4billion.
ForthisreportweareseparatingoutthecategoryofHRsoftware,owingtobothitsevolutionandgrowth.
AnanalystofHRtechnologywritingrecentlyinForbesreferredtothisindustryasbeinginastateof"disruption…andreinvention…fueledbymobileapps,analytics,video,andafocusonteam-centricmanagement.
"101Softwareinthiscategoryfacilitatestheemployeelifecyclefromrecruitmenttoretirement,includingskillsupgradingandwork/lifebalance,andthemoreinstrumentalfunctionsoftaxfiling,payrolladministration,andbenefitsprocessing.
Anoteworthychangeinthisindustryistheshiftfromcomplexandexpensivein-houseHRsoftwaresystemsthatwerebasedonclient/serverarchitecturestoamodelbasedoncloudcomputing,cross-platformfunctionality,andthewidespreaduseofsuchfeaturesasvideoandgamification(e.
g.
theawardingofpointsandbadgesforspecificbehav-iors.
)1026.
6ProductivityToolsThissegmentencompassesservicesdeliveredonlinethatassistbusinesseswithbothB2BandB2Ctasks,rangingfromprinting,graphics,andfilebackuptoaccountingsoftwareandpackagedelivery.
Weusetheterm'ProductivityTools'forthiscategory.
OurestimateforProductivityToolsis$5.
6billionand14,091jobs.
Sage$4061,455Paylocity$2301,800Paycom$2251,461Ceridian$198830Ellucian$44822016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees101http://www.
forbes.
com/sites/joshbersin/2016/07/18/the-hr-software-market-reinvents-itself/#3003447a4930102http://www.
forbes.
com/sites/joshbersin/2016/07/18/the-hr-software-market-reinvents-itself/88Table6.
6:ProductivityToolsAmongthefirmsanalyzedindividually,wenote:Intuit's10-Kfilingindicatesthatthecompany'sConnectedServicessegment(i.
e.
internet-dependent)isresponsibleforgenerating73percentof2015revenue,com-paredwith50percentin2008.
Intuitreportedthat95percentof2015revenuesof$4.
2billionoriginatedintheU.
S.
Dropbox:TheSanFrancisco-basedcompanyenablesfilehostingandgroupcollabo-rationinthecloud.
Whetheratwo-persongrouporalargeorganization,filehost-ing,synchronization,storage,andlivecollaborationacrossgeographiesnolongerrequireinvestmentinsoftwareandhardware.
Thoughprivatelyheldandthereforenotdisclosingfinancials,itsannualrevenuesaresaidtobenear$0.
5billion.
1036.
7GovernmentServicesThesectorofinternet-dependentfederalgovernmentservicesexistsinauniquecategoryinourstudy.
Itdoesnotgeneraterevenuebutitdoescreatesubstantialemployment.
OurestimateforGovernmentServicesjobsis327,440.
Inour2012studyweestimatedapproximately2,000internet-dependentfederalgov-ernmentservicesjobs.
ToarriveatthatestimatewesingledoutthemostpopularandhighprofileU.
S.
governmentsitessuchasCensus.
Gov,USA.
gov,andWhitehouse.
gov.
Intuit$2,9075,340Cimpress(formerlyVistaPrint)$7813,520Web.
com$5432,200LifeLock$587788Dropbox$240720Carbonite$137623SourceMedia$131530Shutterstock$149220SuperMediaInformationServices(superpages.
com)$991502016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees103http://www.
forbes.
com/sites/miguelhelft/2015/07/29/dropbox-is-under-siege-but-its-not-slowing-down/#98849592532589TodaytheU.
S.
governmenthasamuchmorecomprehensivedigitalstrategy,withmanymoresitesanddigitalresources,asreportedonthewebsiteoftheU.
S.
Government'sChiefInformationOfficer:"Overthepastfewyears,theAdministrationhaslaunchedanumberofOpenDataInitiativesaimedatscalingupopendataeffortsacrosstheHealth,Energy,Climate,Education,Finance,PublicSafety,andGlobalDevelopmentsectors.
TheWhiteHousehasalsolaunchedProjectOpenData,designedtosharebestpractices,examples,andsoftwarecodetoassistfederalagencieswithopeningdata.
Theseeffortshavehelpedunlocktrovesofvaluabledata—thattaxpayershavealreadypaidfor—andaremakingtheseresourcesmoreopenandaccessibletoinnovatorsandthepublic.
"104Toestimatethenumberofinternet-dependentjobsintheU.
S.
GovernmentweconsultedthePresident'sIT(InformationTechnology)budgetfor2016.
105Totalexpenditureoninternet-dependentITsystemsandserviceswasgivenas$81.
68billion.
Wedividedthisfigureby$250,000perhead,ourestimateofcost(laborandequipment)peremploy-ee,toarriveattheestimateof327,440jobs.
104https://playbook.
cio.
gov/105https://www.
whitehouse.
gov/sites/default/files/omb/assets/egov_docs/omb_presidents_it_budget_for_fy_2016_summary_chart.
pdf90Untilthispointthenarrativeofthegrowthoftheinternethasbeentoldasifitwasunfoldinginfourindependentchapters,eachcoveringahorizontallayer,frominfrastructuretoconsumerservices.
Asanorganizingprincipleithasworkedreasonablywell,andhasservedourpurposesince2008.
However,itisanapproximation.
Sincetheinternet'sbirththerehavebeenfirmsthatdidnotfittidilyintojustoneofthelayers.
Fromtheearliestdaysoftheinternettherehavebeenattemptstobreakfromthehori-zontallayerstructureandcreateverticallyintegratedecosystemswithinwhich,itwashoped,usersoftheinternetwouldfindrichanddiversecontentandservicesdeliveredundercontractbytheirInternetServiceProviders.
Theseintegratedfirmswerecalledportals.
Yahoowasaportalinthelate1990s,aswereExcite,MSN,andLycos.
Eachacquiredcontentsiteswiththegoalofincreasingthetimethatconsumersspentinsidetheirportalbecauseitcateredtotheirwantsandneeds.
Themostsuccessfulofthepor-talswas,forawhile,AmericaOnline,anditsbriefmergerwithTimeWarnerwasun-dertakenin2000intheexpectationthatthisverticalintegrationwoulddeliverusersanimprovedexperiencewheretheywouldfindengagingcontentandtoolsunderoneroof.
EventuallyGooglecontributedtothedisintegrationoftheAOL/TimeWarnerecosystembypositioningitselfasapuresearchtoolwithoutspecialistcontent.
Inour2012study,wedealtwiththefirmsthatdidnotfittidilyintoonelayeroranotherbydividingtheinternet-dependentrevenuesofthesefirmsamonglayers.
However,by2016thefirmshadgrownandwefounditdifficulttoaccuratelydividetheircontribu-tionsamongdiscretelayers.
Wechosethereforetoexaminethesefirmsontheirownterms,sothatwecouldtakeaviewonwhetherthisreversiontotheverticallyintegratedbusinessmodelwasagainatransitorypatternorthistimearobustfeatureoftheinternetandanexpressionofcon-sumerpreferences.
Thefirmswestudyinthischapterarelistedinthefollowingtable.
Wefoundthattheinternet-relatedemploymentoftheintegratedfirmswas442,000,whichisabout11percentoftheinternetemploymentbase,andrevenuewas$273billion.
Chapter7:IntegratedFirms91Table7.
1:IntegratedFirmsThepatternsofintegrationarenotidentical.
Weseetwomaingroupings:contentmar-ketersintegratingintotransmission,andtransmissioncompaniesacquiringcontent,andathird,themoreidiosyncraticintegrationsofMicrosoftandAmazon.
7.
1ContentMarketersIntegratingVerticallyintoTransmissionTheintegrationpathfromcontenttotransmissionismostclearlyseenintherecentactionsofFacebookandGoogle.
Facebookhasgrownglobalrevenuesbymorethananyotherinternet-dependentfirminthetimesinceourpreviousstudy,from$3.
7billionin2011to$17.
9billionbytheendof2015.
TheusernumbersbehindFacebook'ssuccessaresimilarlyimpressive,reaching1.
5billioninmonthlyactiveusersbytheendof2015.
Constantlyrefininganddevelopingtheircoreproduct,Facebookalsokeepsitseyefixedonstartups,whethertheirfocusisasophisticated,innovativetechnologysuchastheOculusvirtualrealityoffering,orsomethinglesstechnicallycomplex,suchasInstagram,whichquicklycapturedtheimaginationofusersworldwide.
WhenacquiredbyFacebookin2012thephoto-snappingapphadunder100millionusersbutgavethesocialnetworktheworld'slargestrepositoryofphotographs.
Bytheendof2015,aftertwoyearsintegratedwithFacebook,theInstagramuserbasereached500million.
Oth-erFacebookacquisitionsofnotebetweenthetimeofourlastreportandthisoneincludethe2013acquisitionofAtlas,theadvertisingtechnologysuiteownedbyMicrosoft,andthe2014purchaseofWhatsApp,theworld'smostpopularfreemessagingapp.
ThisactivityonitsowndoesnotqualifyFacebookfortreatmentasanintegratedfirmbecausetheseacquisitionsareallextensionsofitspublisherrole.
Buttohouseitsphotostock,whichgrowsbyhundredsofmillionsdaily,itsvideostock,anditsplatformservicesAT&T$65,860126,269Verizon$65,57688,534Amazon$57,337122,324Alphabet(Google)$34,80028,434Comcast$20,82139,084Microsoft$15,98422,200Facebook$8,5135,965CoxEnterprises$4,5009,4082016U.
S.
InternetRevenue($M)2016U.
S.
InternetEmployees92formorethanamillionwebsitesandhundredsofthousandsofapplicationsthatusetheFacebookConnectplatform,ithashadtobuildstorageinfrastructureinfourverylargedatacentersandhastwomoreunderconstruction.
106Inadditionithasjoinedaninvest-mentconsortiumtolaythePacificLightCableNetworkfromLosAngelestoHongKong,andhelpedbuildatransatlanticcablefromBilbaoinSpaintoVirginiaintheU.
S.
107Googlehashadalongerhistoryofexpansionintotransmission,startingwithabidonwirelessspectrumin2008.
ItcreatedtheAccess108divisionofAlphabetin2015tohouseinitiativessuchasthemanufacturingofa$200homerouter,aprojecttopartnerwithanIndianpublicbroadbandproviderknownasRail-Tel,andfiberinstallationsinKenyaandGhana.
ThelargestventureisknownasGoogleFiber,providingbroadbandinternetandcabletelevisionaccessineightU.
S.
cities.
Costdifficultieshaveledtoahaltinlayingoffiberopticcabletohomes,andasearchforahigh-bandwirelessalternative.
7.
2TransmissionProvidersIntegratingVerticallyintoContentTherehavebeensomevigorousmovesbycableoperatorsandphonecarriersintoareasofcontent.
Twocableoperators,ComcastandCox,andtwotelecommunicationscompanies,VerizonandAT&T,haveembarkedonarangeofacquisitionsintendedtodiversifyrevenuesources.
ComcastisthelargesthomeinternetserviceproviderintheUnitedStatesbyvirtueofoperatingthecountry'slargestcablenetwork.
Thecableinfrastructure'snear-exclusivecontroloftransmissionofentertainmentisbeingchallenged.
Consumersareincreasinglyabletounbundlecable'spreviouslyveryprofitablechannelbundles,inparticularbybuyingaudio,video,andothermediaovertheinternetdirectlyfromcontentpublishers,usingover-the-topdeliveryservicesthatgivesISPs,includingcableoperatorslikeCom-cast,noabilitytoshareintherevenueofthecontenttheytransmit.
Anticipatingadeclineinthevalueofitsinfrastructurerelativetothedigitalcontentthatitcarries,Comcasthasrepeatedlytriedtobuycontentproviders.
ItbidforDisneyunsuc-cessfullyin2004,butin2011itcompletedtheacquisitionofNBCUniversal,givingitasignificantpresenceinmanyinternet-dependententitiesincludingashareinHulu.
Verizon,thelargestwirelesscommunicationserviceproviderintheU.
S.
,haspursuedasimilarpatternoflarge-scaleintegrationintoconsumerservices.
ItacquiredAOLin2015andislikelytocloseonYahooin2017.
106DataCenterKnowledge,September2016http://www.
datacenterknowledge.
com/the-facebook-data-center-faq/107Wong,J.
,Quartz.
October17,2016http://qz.
com/811032/google-goog-and-facebook-fb-are-building-a-new-transpacific-submarine-cable/.
Retrieved11/8/2016.
108Bergen,Mark,"MeetAccess,theGoogleUnitThat'sTakingOnComcastandtheRestoftheCableBiz,"Recode.
November30,2015http://www.
recode.
net/2015/11/30/11620972/meet-google-alphabets-access-and-ener-gy-division-home-to-google-fiber93CoxEnterprises,aprivatelyheldcorporation,hasfollowedadifferentpathtocontent,concentratingitsfocusontheautomotivesector.
ThroughCoxAutomotiveitownsanumberofautomotivecontentpropertiesincludingKelleyBlueBookandAutotrader.
com,andconnectedservicesintheautomotivevertical,inparticularDealertrackwhichprovidesonlinefinanceandinsuranceservicestodealers.
AT&T'sacquisitionofDirecTV,satelliteinfrastructureprovider,isnotperseasteponthepathtointegration,butifitsoffertoacquireTimeWarnerissuccessfulitwillhavemadeoneofthelargestmovestotieinfrastructuretocontent.
Andevenifitisnotsuccess-ful,AT&ThasasubstantialstakeintheconsumerservicessupportlayeroftheinternetthroughAT&TAdWorks,whichsincetheDirecTVacquisitionhasgrowntoanannualrevenueof$1billionindigitalandcross-deviceadvertisingsales.
7.
3IntegratingVerticallyintoCloudTransmissionandDataServicesMicrosoft'smultiplerevenueandemploymentgroupingswouldhavebeenmoredifficulttoclassifythanperhapsanyotherfirminouruniversehadwenotcreatedanIntegratedFirmslayer,andevenwiththislayeritspatternofdiversificationisrelativelysuigeneris.
Itreportsrevenueinsixsegments,twoofwhichloadtosomedegreeontotheinternet.
Thefirst,Devices&ConsumerLicensingcontainsthesearchengineBing,andgamingsoftwareanddevicesthatrangefromthesubstantiallyinternet-dependentMinecraftsoftwareandtheinternet-accessibleXboxhardware,tosomenon-internetdependentgaming.
Italsocontainsthephonedivision,whichappearstobedeclininginstrategicimportance.
Thesecond,CommercialLicensing,containsSkype,whoserevenuesarenotseparatelyreported.
Weassume$2.
5billionfroma2013reportwithaconservative10percentgrowthperyear.
109Thissegmentincludesthesuccessfulcloudentry,Azure,anditscustomerrelationshipmanagementtools.
ThesignificanceoffirmslikeMicrosofttothegrowthoftheinternetliesperhapsinthepaththeypointtoforenterpriseinformationtechnologyintegration.
Firmsthatcanbuildonlongandstronghistoricalrelationshipswithlargeenterprisescanpioneercloud-basedbusinessmarketsfortheirclients.
InthisrespectMicrosoft's$26.
2billionacqui-sitionofLinkedIn,intersectingitwithSalesforceandWorkday,pointsuparolefortheinternetinbusinesslifeaslargeassocialmedia'sroleinpersonallife.
Amazonisthelastfirmexaminedinthischapter,andlikeMicrosoftitspathtointegra-tionissuigeneris.
Amazon'scorebusinessise-commerce,whereitisdominantandincreasingindominance.
Itisusefultoreviewherethee-commercepictureandAma-zon'splaceinit,becausee-commerceaccountsonitsownforabout20percentoftheemploymentintheentireinternetecosystem.
Therearemorethan500retailerssellingonlineintheUSA,andalthoughofflinesalesare92percentofallU.
S.
retailtodaytheonlinesectorisgrowing.
Remarkably,Amazonsellsafifthofallthatissold,andits109http://www.
bloomberg.
com/news/articles/2013-02-19/microsoft-s-skype-unit-approaching-2-billion-in-annual-revenue94shareisgrowingnotdecliningasthesectorgrows.
Italonecaptured60percentofallgrowthin2015.
Thisunusualpattern,inwhichthelargestfirminanindustrygrowsandincreasesitsshareasitgrows,suggestsincreasingreturnstoAmazon'sscale.
Thelargeritgets,themorestronglyitcompetes,likelybecauseitstechnologyinvestmentsyieldhigherreturnsthanthoseofcompetitors.
Amazon'sintegrationhasbeenfromtheconsumerservicelayertotheconsumerservicessupportandsoftinfrastructurelayers.
Byofferingitstechnologyinvestmentsasinfrastruc-turetothemarketgenerallyithasgrownitsscaleandtheattendantbenefitstoitscorebusiness.
Andbyofferingadvertisingandcustomermanagementservicestothird-partyvendorssellingonitsplatform,ithassimilarlygrownscaleinconsumerservicessupport.
7.
4IntegrationConclusionsGoogle,Facebook,andAmazonprovideplatformservicestoclients,mainlyretailers,publishers,oradvertisers,whichinsomeinstanceslimitaccessandcontroloverclientfirms'applications,content,andmediawhentheyrunontheirservices.
Thedatagen-eratedbytheirservicesistheirfirst-partyproperty,notthepropertyoftheclientfirms.
Inexchangeforthebenefitsoftheseecosystems,theclientsearnareturnontheirinvest-mentsintheformofretailsales,advertisingrevenues,andadaudienceexposuresthatarecompetitivewith,andoftenbetterthan,thereturnstobeearnedintheopeninternet,whileprovidingmorecontrolandsafeguardsoveruserexperience.
Asof2016theseecosystemsareextremelysuccessfulandareexperiencingtremendousgrowth.
Itappearsthatthecableprovidersandtelecommunicationscompaniesreviewedearli-erinthischapterareaccumulatingassetswiththegoalofbeingabletooffersimilarlycompleteservicestomarketers.
Somedependonregulatoryapproval,andothersontheabilitytoacquirecomplementarybusinessesatreasonableprices,butitseemslikelythatallofthefirmslistedherewilleventuallyoffersuitesofmarketingservicesaswell.
WilltheseecosystemsbemoresuccessfulthantheverticallyintegratedcompaniesofthepastFirstitshouldbeclearthat25yearsagoitwasconsumerswhowerecaptivetotheseecosystems,whiletodayitismarketingfirms.
Butmoregenerally,theecosystemsarenotmonopolies.
Marketingfirmscananddoruncampaignsinmultipleenviron-ments,andallocateresourceswheretheyseethebestresults.
Providedobjectivemea-suresofresultsareavailable,integratedfirmswillonlythriveiftheydeliverbetterresultsthancombinationsfromamongtheopeninternet.
95Inthischapter,wecombinethefindingsofthefivelayersoftheinternetecosystemintoawhole.
Thefivepreviouschaptersanalyzedemploymentintheinternetecosystembyidentifying412ofthelargestfirmsforindividualanalysisin48relativelyhomogeneousclusters.
Additionally,weidentified23groupsoffirmsorself-employedindividualsforcollectiveanalysis.
Ofthese,10were"allother"clusters,generatedwheneverthedis-tributionoffirmsbysizeinoneofthe48clustersleftasignificantnumberofuncountedfirms.
Anothernineclustersaccountedforself-employedworkerssuchassellersonEtsyandindividualsellersoneBay,on-demandeconomyworkers,andfreelanceindividualsdoingcoding,contentcreation,andotherservicesforWebsites.
Wemadeallowancesintwocasesforsectorscomprisingfirmsthatwereindividuallysmallbutlargeinaggre-gate,namelylocalandstatee-governmentservices,e-commercemerchantsbeyondthetop10,andpeopleworkinginsmallunitsoflargeandmid-sizedgeneralenterprisesnototherwisecountedintheinternetecosystemwhoowedtheirjobstotheinternet.
8.
1EmploymentTotaldirectemploymentintheinternetecosystemis4,097,001,ofwhich2,313,000camefromthe412individuallyenumeratedfirms.
Therefore,ourenumerationoflargefirmsproduced56percentofthe4.
1millionpeopledirectlyemployedintheinternetecosystem.
Theremainderworkedinmid-sizedandsmallfirmsorwereself-employedindividuals.
Totalemploymentisdistributedasfollowsacrossthefivelayersoftheeco-system,withemploymentfor2008and2012shownforcomparison:Table8.
1:DirectEmploymentintheInternetEcosystemChapter8:ConclusionsRegardingEmploymentintheInternetEcosystemInfrastructure/HardInfrastructure140,000420,000304,393InfrastructureSupport/SoftInfrastructure165,000254,000662,691CustomerServicesSupport190,000435,0001,068,364ConsumerServices520,000885,0001,619,335IntegratedFirms442,218TOTAL1,015,0001,999,0004,097,001GrowthinEmployment(%perannumcompound)18.
5%19.
6%2008EmploymentLayer2012Employment2016Employment96Thusthe2016bottom-upanalysisfindsthatthenumberofjobsthatrelyontheU.
S.
internetecosystemhasdoubledeachfouryears.
Almostamillionnewjobswereaddedbyearly2012tothemillionthatwerefoundinthe2008study,andalittlemorethantwomillionwereaddedbylate2016tothetwomillionwerefoundinthe2012study.
Thuswefindthatthenumberofpeopleemployeddirectlyintheproductionofservicestotheinternetecosystemisabout4.
1million.
Foreachpersondirectlyemployed,otherpeopleworkinsectorsthatservicetheneedsofthisperson,suchasschooling,entertain-ment,banking,insurance,andretail.
Thepersonpaystaxesthatsupportemploymentinfederal,state,andmunicipalgovernmentservices,education,andthemilitary.
Thisindirectemploymentarisesfromsuppliereffects,re-spendingeffects,andgovernmentemploymenteffects.
Itisstandardpracticetoapplyamultipliertothedirectemploymenttoaccountfortheindirectemploymentthatwouldbelostifthedirectemploymentdidnotexist.
TheU.
S.
BureauofLaborStatistics(BLS)publishesstatisticsonindustryem-ploymentrequirements,whichenablecalculationofthesemultipliers.
Sectorsdifferinthesizeoftheirmultipliers.
Bivens110hascomputedindirectemploymentmultipliersthatrangefrom372indirectjobsforevery100jobsindurablesmanufacturingto163indi-rectjobsforevery100jobsinbusinessservices.
Theseestimatesareinclusiveofcapitalserviceusage.
Hann,Viswanathan,andKoh111usedarangeofmultipliersfrom2.
4to3.
4intheiranalysisoftheFacebookappeconomy,whileMandelused0.
5inhisreportontheappeconomy.
Wehavechosenaconservativemultiplierof1.
54,theratiousedinourearlierreports.
Thusourprojectionofemploymentduetotheadvertising-support-edinternetecosystemis4.
1milliondirectjobsand6.
3millionindirectjobs,foratotalemploymentof10.
4millionpeople.
SectorcontributiontoGDPisthesumofnationalincome,corporategrossprofits,andinterest.
Aswasdonein2008and2012,thesectorcontributionwastakentobeproportionaltosectorincomeusingthenationalaveragewageindexpublishedbytheU.
S.
SocialSecurityAdministration.
Weusedawageindexof$48,880andapropor-tionalfactorof2.
21forconsistencywithpastreports,togiveacontributionperdirectandindirectjobof$108,000.
Bythismethod,employmentintheadvertising-supportedinternetecosystemcontributedabout$1,121billiontotheU.
S.
GDP.
Applyingthesameprocedureto2008and2012employment,wegeneratethecontri-butionsoftheinternetecosystemtoGDPinprioryears:110Bivens,J.
,2003.
"UpdatedEmploymentMultipliersfortheU.
S.
Economy.
"EPIWorkingPaperNo.
268.
111Hann,IH,Viswanathan,S.
,Koh,B.
"TheFacebookAppEconomy.
"CenterforDigitalInnovation,UniversityofMaryland,201197Table8.
2:ContributionoftheInternetEcosystemtoU.
S.
GDP8.
2NewPhenomenaThisstudy,concernedwithwhatis,risksneglectingwhatwillbe.
Intheselectionoffirmsandclustersoffirmsourattentionwasinevitablydrawntothelargesitesofvalue.
Soinourpreviousstudies,forexample,themobileinternetbarelyratedconsiderationin2008butby2012wasbecomingsignificant.
Smartphonesbegantooutsellpersonalcomputersin2011.
Inthisreport,theinfluenceofthemobileinternetispervasiveateverylevelfromhardinfrastructuretoconsumerservices.
Omittedaretwonewphenomenawhichare,asmobilecomputingwasbefore,onthefringesofthereportbutnotyetlargesitesofvalueoremployment.
ThefirstistheInternetofThings(IoT).
Thetermisusedtodescribeobjects,someloca-tion-boundandsomemobile,thatarenetworked.
Theobjectscanrangefromdevices,tovehicles,tobuildings,eventoelementsofthehumanbody,andtheybecomepartoftheIoTwhentheyareembeddedwithsensors,actuators,software,andnetworkconnec-tivitysothattheycollect,transmit,andexchangedata.
IoTisacoreplatformininitia-tivesbrandedasSmartCityandSmartEnergyManagementSystems,aswellasmoremundaneapplicationssuchasmediaaudiencetracking,fleetvehiclemonitoring,andhomeautomation.
ThepresenceofIoTisunderstatedinthisreportfortworeasons.
Firstmanyinfrastructur-alelementsandapplicationsarebuiltonmobilityanduseitsinfrastructure.
FirmsthataredominantintheInfrastructurechaptersalreadyearnsomeoftheirrevenuefromIoTproductsandservices.
ExamplesareCisco,Intel,Qualcomm,Ericsson,andMicrosoft.
Othersnotaslarge—suchas,LogMeIn,RedHat,andNovatel—earnrevenueforIoTsoftwarenotseparatelyidentifiedassuchinourtables.
Second,manyapplicationsaretoosmalltodaytomeetourminimumthresholdforinclusion.
112ThesefirmsincludeDirectemploymentduetointernet1,015,0001,999,0004,097,001Directandderivedemployment3,050,0005,100,00010,383,000ContributionofinternettoGOP$300billion$530billion$1,121billionGrowthinGDP(%perannumcompound)15.
5%20.
0%ShareofTotalUSGDP2.
1%3.
7%6.
0%2008Report2012Report2016Report112Seehttp://www.
postscapes.
com/companies/foradetailedlistofIoTcompanies.
98Samsara,Notion,Hologram,andLosant,pioneersinIoThardwaresuchasGainspace,Particle,andLibelium,softwareproducerssuchasProSyst,Litbit,andAntmicro,andcloudvendorssuchasAylaNetworksandXively.
AndtherearesystemsintegratorswhospecializeinIoTsuchasAmyx,ThinkLogixandFlex.
SowhilethereisnobreakingoutofIoT-sourcedrevenuesinthisreport,someiscaughtinthenetoffirmsthatweanalyzed.
ItisinevitablethatIoTwillimposeaverysignificantloadontheinternetoverthefiveyearstocome.
Asecondnewphenomenontoosmalltodaytowarrantmuchattentioninthisreportisartificialintelligenceanditsconcretemanifestation,bots.
Thephenomenonisnotnew,butsuccessfulimplementationsare.
Microsoft'sdancingpaperclip,apersonalizationofartificialintelligencetoaddressandanticipateproblemsinusingtheMicrosoftsuiteofofficeproductivitytools,wasanearlyfailure—Clippythepaperclipwasannoyingandcounterproductive.
Todaybotsarebeginningtobewelcomed,andusedforeverythingfromsimplecustomerservicequeriestocomplexfinancialtransactions.
FamiliarasApple'sSiri,botscombineaconversationalinterfacewithartificialintelli-gencecapabilities.
Botsarenowenteringeverydaylife.
TheNewYorkTimesusesonebuiltwithinSlack,acloud-basedteamcollaborationtool,todetectwhicharticlesarebeingviewedandsharedatunusuallyhighratesandthereforearelikelytogoviral.
TheSiliconValleystartupTrimhasbuiltabotthatcancelsunwantedrecurringsubscriptionsandhasevenbeenprogrammedtoarguewithComcast'scustomerservicerepresenta-tivestonegotiatealowerbill.
113TheO'Reillytechnologynewsletterexplains,"Agoodbotisamarriageoflow-frictioninterfaceandartificialintelligence.
Theinterfaceletsauseraskaquestion,stateafact,orexpressawishintextorspeech.
TheAItakesfamiliarhumanlanguage,extractsinformationfromit,organizesit,andactsonit.
"114Thetwotopicsonthefringesofthisreport,botsandtheIoTarenotindependentinmanyforeseeableapplications.
AutonomoustransportationforexamplewouldrequirethemarriageofAIandIoT.
Afleetofself-drivingvehicles,suchasconstructionorearth-movingvehiclesalignedonasingletask,mightbeconstruedasasingleentity,arobotextendingacrossdistance,eachvehicleaclientandaserver.
113http://fortune.
com/2016/11/16/trim-comcast-bot/114https://www.
oreilly.
com/ideas/why-2016-is-shaping-up-to-be-the-year-of-the-bot)99Chapter8findsthattheadvertising-supportedinternetecosystemcontributesabout$1,121billiontotheU.
S.
GDPin2016.
Wenowattempttocorroboratethiscon-clusionfromtwounrelatedperspectives.
9.
1TheInternet'Exports'totheRestoftheEconomyInthismethodwefirstestimatethedirecteconomicvalueoftheservicesthattheinternetprovidestotherestoftheU.
S.
economy,whichistherevenuepaidfortheservicesper-formedfortherestoftheU.
S.
economybeyondthebordersoftheinternetecosystem.
Theserviceswereofthreekinds.
First,theecosystemexportedadvertising.
Thesumofsearch,display,classified,mobile,leadgenerationandemailadvertisingservicesinthelasthalfof2015andthefirsthalfof2016was$65billionaccordingtotheinteractiveAdvertisingBureau.
115Second,theecosystemexportedretailservices.
Retailrevenuesareestimatedinthisreportat$604billion,andwithanestimatedgrossmarginof50percentforcostofgoodstheserviceproducesvalueof$302billion.
Third,theecosystemprovidedinternetaccess,forwhichitchargedtherestoftheeconomy,asestimatedinthisreportbysummingthetransmis-sionandconnectivitysegmentsoftheHardInfrastructurelayer,at$210billion.
Thustotalnetexportswerethesumofthesethreepayments,$577billion.
Totranslatetheecosystem'spaymentsforservicesintoasectorGDPweappliedanin-directemploymentmultiplierof1.
54.
Bythismethod,theadvertising-supportedinternetecosystemcontributesabout$888billiontotheU.
S.
GDPin2016.
9.
2TimeSpentontheInternetFirst,theecosystemexportedadvertising.
Thesumofsearch,display,classified,mobile,leadgenerationandemailadvertisingservicesinthelasthalfof2015andthefirsthalfof2016was$65billionaccordingtotheinteractiveAdvertisingBureau.
115Second,theecosystemexportedretailservices.
Retailrevenuesareestimatedinthisreportat$604billion,andwithanestimatedgrossmarginof50percentforcostofgoodstheserviceproducesvalueof$302billion.
Third,theecosystemprovidedinternetaccess,forwhichitchargedtherestoftheeconomy,asestimatedinthisreportbysummingthetransmis-sionandconnectivitysegmentsoftheHardInfrastructurelayer,at$210billion.
Thustotalnetexportswerethesumofthesethreepayments,$577billion.
Chapter9:Corroboration115http://www.
iab.
com/wp-content/uploads/2016/04/IAB_Internet_Advertising_Revenue_Report_HY_2016_Final-POSTED.
pdf116http://totalaccess.
emarketer.
com/Results.
aspxdsNav=N:1240117http://www.
bls.
gov/news.
release/pdf/atus.
pdf100TheeMarketerreportfindsthatpeopleover18intheU.
S.
spend310minuteseachdayonallinternetactivitiesondesktopcomputers,laptopcomputers,andmobiledevices.
Notethatthisanalysisreportsthesumofminutesspentwitheachmedium,withoutanyattempttoaccountforsimultaneoususeofmorethanonemedium.
Ourneedhereisforthefractionofthedayspentinsomeformofonlineactivity,sowemadeasubjectiveallowanceforduplication.
TheU.
S.
populationovertheageof18is240million.
WehaveestimatedthevalueofanhourspentatworkforarepresentativeU.
S.
workerat$23.
50perhour,basedonannualearningsof$48,880derivedfromtheaveragewageofnon-management,non-agriculturalworkerspublishedbytheBureauofLaborStatistics118.
Thereisnomarketpriceforanhourspentinrecreationorleisure,althoughthereisanopportunitycost.
IthasbeenarguedbyBockstaeletal.
119thatthewageratemeasurestheopportu-nitycostofleisuretime.
Ifnot,thewagerateover-estimatesthevalueofaleisurehour.
Asanapproximation,weuse10percentofthewagerateforleisuretime,andwetake75percentofonlinetimeasleisuretime.
Withtheseassumptions,310minutesperdayincursanopportunitycostof$3,451billionannually.
But,asjustnoted,notallofthisamountistrueopportunitycost,assomeinternetuseoccursinparallelwithotheruse,aswheninternetbrowsingoccurswhilelisteningtostreamedmusicinbackground,andprivateuseoccursinparallelwithworkplaceactivity.
Wetakeduplicativeusetobetwothirdsofalluse,soweadjustourestimateofthecontributionoftheinternettoGDPbythismethodto$1,150billion.
9.
3ComparisonoftheMethodsThuswehavethreeestimates:$1,121billionbymethodone,$888billionbymethodtwo,and$1,150billionbymethodthree.
Thepurposeofthesecondandthirdmethodsis,asstatedinthemethodologychapter,toprovidesomedegreeofindependentcor-roborationforthefirstmethod,and,giventhattheyarereasonablyaligned,webelievemethodonehasbeencorroborated.
118http://www.
census.
gov/compendia/statab/cats/labor_force_employment_earnings.
htmlaccessedAugust24,2012.
119Bockstael,N.
,Strand,I.
,andHanemann,W.
(1987).
TimeandtheRecreationalDemandModel.
AmericanJournalofAgriculturalEconomics.
69(2),293-302.
101ThischapterisintendedtoshowthedispersionofeconomicbenefitsoftheinternetecosystemacrosstheU.
S.
,andthechangeindispersionsince2012.
10.
1MethodologyU.
S.
FederalstatisticalagenciesusetheNorthAmericanIndustryClassificationSystem(NAICS)tocategorizebusinessestablishmentsforthepurposeofcollecting,analyzingandpublishingdata.
BydetailedreviewofNAICSdefinitions,weconcludedthat15oftheapproximately700five-digitNAICScodescontainmostoftheemploymentintheinternetecosystem.
WethenreliedontheU.
S.
CensusBureau'sCountyBusinessPatterns(CBP)datasettodeterminethecountiesinwhichtheemployeesinthoseNAICScodesworked.
TheCBPreportsthenumberofemployeesbyNAICScodeandcountyinestablishmentslistedintheCensusBureau'sBusinessRegister,adatabaseofallknownsingleandmulti-estab-lishmentemployercompanies,butexcludesamongothersPostalServiceemployees,privatehouseholds,publicadministration,andmostestablishmentsreportinggovern-mentemployees.
Fortheseexclusionswereliedonourbottom-upestimatesofjobsanddistributedthemacrosscountiesinproportiontothesizeofthecountyasdiscussedlater.
TheCensusBureauisnotpermittedbylawtodiscloseinformationthatidentifiesthenumberofpeopleemployedataparticularestablishment.
Thereforeinthelimitednumberofcaseswhereitiseasytoinferanestablishment'sidentitybecausetherearesofewestablishmentsinacounty,orbecauseoneestablishmentinacountyisconspic-uouslymuchlargerthantheothers,theCPBreportsthenumberofestablishmentsbutnotemploymentforthatcounty.
Whenemploymentisnotreported,itisreferredtoassuppressed.
Ineachofthe15NAICScodesstudiedherewereplacedacounty'ssup-pressedemploymentbyanestimatecalculatedasfollows.
ForeachNAICSwehavethesumofnationwideemployment,andcancompareittothesumofallunsuppressedcountyemployment,yieldingthenationwidetotalforsuppressedemploymentforthatNAICScode.
Weallocatethesuppressedemploymenttocountiesinproportiontothenumberofestablishmentsinthatcountywhoseemploymentwassuppressed.
Chapter10:DistributionofInternetEmploymentbyGeography:Top-DownApproach102AsTable10.
1.
1shows,suppressionwasnotsignificantinthelargestofthe15NAICScodesthatwestudied,sothatmostoftheemploymentthatwereportedwassourceddirectlyfromtheCPB,exceptintheexcludedcategoriesnotedabove.
Table10.
1.
1:SummaryofSelectedNAICSCodesFromourshortlistof15relevantNAICScodes,wedeterminedthattwowereentirelyinternet-dependent,andtheremaining12werepartiallyinternet-dependent.
TodeterminethefractionofinternetemploymenttousewhenaNAICScodewaspar-tiallyinternet-dependent,weusedoneoftwomethods.
Ifpossiblewereliedondatacompiledfromourbottom-upanalysis,usingcompanies'internetrevenueandtotalU.
S.
541510Computersystemsdesignandrelatedservices244,2661,917,8052%541610Managementconsultingservices183,288995,2962%541800Advertising,publicrelations,andrelatedservices49,767479,6493%523120Securitiesbrokerage23,217277,21312%517110Wiredtelecommunicationscarriers19,834587,79215%454111Electronicshopping18,844196,60820%518210Dataprocessing,hostingandrelatedservices17,933301,56910%511210SoftwarePublishers17,411332,2716%519130Internetpublishingandwebsearchportals12,997186,1125%492110Couriersandexpressdeliveryservices10,374547,19422%517210Wirelesstelecommunicationscarriers7,939132,13935%517910Othertelecommunications7,18280,28533%334200Communicationsequipmentmanufacturing2,28688,29629%334100Computerandperipheralequipmentmanufacturing1,721161,34730%454112Electronicauctions75610,37081%N/AEmploymentinindustriesexcludedfromtheCBP**N/A1,116,918N/A*TotalsincludeU.
S.
territories**Manyemployedandself-employedpeoplearenotcountedintheCPBsurvey,discussedaboveas"exclusions.
"NAICSCodeIndustryTitleNumberofEstablishments*TotalEmployment*%ofTotalEmploymentSuppressedattheCountyLevel103revenue,tocalculatean'internet-intensity'ratioforeachNAICS.
Eachcompanyinourbottom-upanalysiswasassigneditsappropriateNAICScode.
Ifwehadarepresenta-tivesampleofcompaniesforaparticularNAICS,wecomputedthe'internet-intensity'ratioandthenestimatedfractionalemploymentfortheNAICSbymultiplyingits'inter-net-intensity'ratiobyitstotalemployment.
Ifourbottom-upanalysisdidnotresultinarepresentativesampleofcompaniesforaselectedNAICS,weutilizedProductLineReceiptsfromthe2012EconomicCensus.
ThisdatacontainsthedollaramountofreceiptscollectedbymanyindividualproductlineswithineachNAICSindustryandalsohighlightsvaluabledetailaboutthetypesofproductsandservicestheNAICSindustriessold.
ForeachoftheselectedNAICSindustries,allProductLineReceiptswerereviewedtoidentifywhichProductLinescouldbeconsideredapartoftheinternetecosystem.
ToestimatefractionalemploymentforaNAICS,wemultiplieditsratioofinternet-relatedProductLineReceiptstoTotalProductLineReceiptsbyitstotalemployment.
Asmentionedabove,theCBPdoesnotassignPostalService,privatehouseholds,andgovernmentservicesemployeestocounties.
Wemakeanallowanceforthisbyincor-porating:(i)30percentofU.
S.
PostalServiceemployment(theproportionofrevenuegeneratedfromtheshippingofe-commercepackagesfromourbottom-upanalysis);(ii)full-timeequivalentemploymentofindividualsellersonsitessuchasCraigslist,Amazon,Etsy,andeBay;and(iii)e-governmentemployees.
Weallocatedthisemploymenttocountiesinproportiontocountypopulations.
Inthismannerwewereabletoassign3.
192millioninternet-dependentemployeestoNAICScodes(Table10.
1.
2),tostates(Table10.
2),andtocongressionaldistricts(Table10.
3.
1).
104Table10.
1.
2:Internet-DependentEmploymentforSelectedNAICSCodesOurtop-downestimateof3.
192millioninternet-dependentemployeesislessthanourbottom-upestimateof4.
1millionemployees.
Thereasonisthatinthissectionwemod-eledinternet-dependentemploymentforonlythemostrelevant15NAICSindustries,thosewithameaningfullevelofinternetactivity,andforwhichwecouldasapracticalmatterestimatefractionalinternetemployment.
TheCBPenumeratescounty-levelbusi-nessstatisticsforover700NAICSindustriesatthe5-digitlevel,accountingfortensofmillionsofemployees.
Hadweexaminedevery5-digitNAICScode,theproportionofinternet-dependentemploymenttototalemploymentinthecodesbeyondthebig15wouldhavefallenrapidly,andthelowertheproportion,thelessconfidencewewouldhavehadinourestimates.
Butacrossthewhole700segments,smallerrorsinestimationwouldhaveaccumulatedtobecomealargeerrorinthefinalanswer.
Inthebottom-up541510Computersystemsdesignandrelatedservices25%477,958517110Wiredtelecommunicationscarriers50%290,311541800Advertising,publicrelations,andrelatedservices50%238,503454111Electronicshopping100%196,227519130Internetpublishingandwebsearchportals100%184,709492110Couriersandexpressdeliveryservices30%163,536541610Managementconsultingservices10%98,837518210Dataprocessing,hostingandrelatedservices30%91,066334100Computerandperipheralequipmentmanufacturing50%80,507511210SoftwarePublishers20%66,333523120Securitiesbrokerage20%55,341334200Communicationsequipmentmanufacturing50%43,909517910Othertelecommunications50%40,255517210Wirelesstelecommunicationscarriers30%39,086454112Electronicauctions100%10,334N/AEmploymentinindustriesexcludedfromtheCBP**100%1,115,127TOTAL3,192,039NAICSCodeIndustryTitleTotalInternetEmploymentIncludingAllocations%ofInternetDependentEmployment105methodologythereareerrorstoo,buttheerrorsarerandomandthereforeself-correct-ing.
Inthetop-downmethodology,theerrorsincreaseasthefractionofinternetemploy-mentintheNAICSsectorsdecreases,andarethereforenotself-correcting.
Weconcludethatthetop-downmethodologyisareasonableestimateofthelowerboundonthesizeoftheinternetecosystem,andareasonablebasisforconservativelyallocatingemploymenttoU.
S.
countiesandcongressionaldistricts.
10.
2ResultsbyStateWeshowinTable10.
2therankingofemploymentbystate,andthegrowthsinceour2012report.
Employmentin2012isbasedonthesame15NAICScodesandthesamerulesforallocationthatwereusedinthisreport.
106Table10.
2:Internet-DependentEmploymentbyStateandDistrictofColumbia1California478,15715.
0%117.
0%2Texas253,0977.
9%116.
9%3NewYork243,0037.
6%113.
2%4Florida181,3495.
7%110.
0%5Illinois132,5164.
2%106.
6%6Pennsylvania109,3823.
4%103.
1%7Washington108,0793.
4%119.
7%8Virginia106,5673.
3%99.
7%9Georgia104,6133.
3%111.
1%10Ohio102,5243.
2%111.
2%11NewJersey95,1313.
0%99.
0%12Massachusetts94,8083.
0%100.
5%13NorthCarolina91,5212.
9%126.
8%14Michigan77,1162.
4%105.
7%15Colorado67,8952.
1%104.
6%16Arizona67,3822.
1%138.
6%17Maryland61,8981.
9%100.
6%18Missouri57,1281.
8%121.
6%19Tennessee55,6191.
7%111.
4%20Minnesota54,5191.
7%107.
4%21Indiana50,5501.
6%107.
5%22Wisconsin45,2461.
4%107.
5%23Kentucky39,0081.
2%109.
0%24Utah38,0831.
2%110.
3%25Oregon36,3391.
1%104.
2%26Alabama36,0331.
1%103.
7%27Connecticut35,8141.
1%93.
5%28SouthCarolina35,2481.
1%118.
2%29Louisiana29,8280.
9%111.
1%30Kansas25,8850.
8%80.
9%31Oklahoma25,7370.
8%90.
9%32Iowa24,3440.
8%103.
8%33Nevada22,1340.
7%102.
1%34Arkansas21,1490.
7%113.
3%35Nebraska18,3100.
6%102.
0%36DistrictofColumbia18,0880.
6%122.
6%37Mississippi18,0120.
6%102.
1%38NewHampshire15,4470.
5%105.
7%39Idaho14,1590.
4%110.
4%40NewMexico13,1640.
4%97.
1%41WestVirginia11,7920.
4%107.
1%42Hawaii10,1000.
3%96.
8%43Maine9,8500.
3%99.
4%44RhodeIsland9,5420.
3%111.
2%45Montana8,8130.
3%98.
0%46Delaware8,1680.
3%107.
7%47SouthDakota7,0930.
2%103.
6%48Alaska6,4020.
2%246.
0%49NorthDakota6,0670.
2%113.
3%50Vermont5,6820.
2%102.
8%51Wyoming4,3060.
1%108.
7%TOTAL100.
0%StateInternetEmploymentGrowthsince2012Report(100%impliesdouble)%ofTotalU.
S.
InternetEmploy-mentStateInternetEmploymentGrowthsince2012Report(100%impliesdouble)%ofTotalU.
S.
InternetEmploy-ment10710.
3ResultsbyCongressionalDistrictAsistobeexpected,asmallnumberofcongressionaldistrictsarehometoahighconcentrationofinternetemployment.
Whatismoreinterestingistheextenttowhichinternetemploymentisdispersedbeyondtheseconcentrations.
Ourcalculationsshownocongressionaldistricthasfewerthan1,296peopledependentontheinternetwheth-erasemployeesorasself-employedpeople.
AndasshowninTable10.
3.
1,theareasofhighconcentrationofinternetemploymentaccountforonlyabout14percentoftheecosystem'stotalemployment.
Table10.
3.
1:DistributionofInternetEcosystemEmploymentbyCongressionalDistrictThetop14congressionaldistrictsareshowninthenexttable:Table10.
3.
2:Top14districtsFirst14congressionaldistricts443,72814%Next102congressionaldistricts+DC1,149,89736%Final320congressionaldistricts1,598,41450%TOTAL3,192,039100%EmploymentPercentCaliforniaCongressionalDistrict1953,873SanJoseNewYorkCongressionalDistrict1343,382UpperManhattanCaliforniaCongressionalDistrict1736,891SiliconValley(Sunnyvale,Cupertino)CaliforniaCongressionalDistrict1234,576CityofSanFranciscoWashingtonCongressionalDistrict733,435SeattleWashingtonCongressionalDistrict932,972Bellevue,TacomaNewYorkCongressionalDistrict1231,429EastManhattanNewYorkCongressionalDistrict1031,240WestManhattanandpartsofBrooklynVirginiaCongressionalDistrict828,281ArlingtonCountyandpartsofFairfaxCaliforniaCongressionalDistrict1824,642PaloAlto,RedwoodCity,MenloPark,etc.
UtahCongressionalDistrict423,967SaltLakeCountyCaliforniaCongressionalDistrict1423,942SanMateo,Burlingame,SouthSFTexasCongressionalDistrict3523,876SanAntonio,AustinMassachusettsCongressionalDistrict521,222EasternMassachusetts,Route128DescriptionEmployment108Figure10.
1plotsemploymentontoU.
S.
congressionaldistricts,andFigure10.
2breaksoutfourdenseregionsofthecountry.
Thecongressionaldistrictsaregradedintodeciles,fromthe10percentofdistrictswithfewestestablishmentstothe10percentwiththegreatestnumber.
Notethatcongressionaldistrictswithlargelandmassesbutsmallpopulationsarevisuallyprominentinthesemaps.
ThusMontana,withonlyonecongressionaldistrictbut8,816internetemployees,isintheninthdecileofthedistributionofcongressionaldistricts,anditsdarkblackshadingmakesitappearmoreprominentthanthestateofWashington,whichhas108,000employeesacross10congressionaldistricts.
ForsimilarreasonsthelightlypopulatedfirstcongressionaldistrictinthestateofNewYork,locatedontheeasternendofLongIsland,isvisuallymoreprominentthanalloftherestofManhattan.
109Figure10.
1:MappingofInternetEcosystemEmployeesontoCongressionalDistricts110Figure10.
2:DetailMapsclockwiseSanFrancisco,Seattle/Tacoma,DCtoBoston,:SanAntonio/Austin111TheimpactoftheinternetismoreprofoundthantheeconomiceffectslaidoutinChapters3through8.
Amultitudeofsocialbenefitsarealsogenerated,feedeco-nomicbenefits,andleadtoavirtuouscycleinwhichconsumers'andcitizen'sneedsaremetbyfor-profitbusinessesandcommunity-focusedorganizations.
Atthesametime,societyhasmovedwellbeyondthenoveltyofthenarrativethatev-eryonecanhaveablog,website,podcast,orYouTubechannel.
Whileitistrueonthesurfacethatexpressionhasbecomedemocratized,functioninginanenvironmentofextremeabundanceofcontentandideasposesanewsetofchallenges.
Bydefault,allvoicesontheinternetcompeteglobally,butalsohavetheagilityandlowstartupcoststhatmakeadjustingorchangingcourserelativelyeasy,andinthiswaysmallscalehasthepotentialtobeusedcompetitivelyagainstlargescalelegacyprocesses,andbyinsurgentactorsagainstentrenchedinstitutions.
Thischapterprovidesanoverviewofthearrayofbusinessmodels,organizationalstructures,andconsumerbehaviorsandpreferencesthatcharacterizethehighlypartic-ipatoryinternetecosystem.
Whatstartedasagovernment-fundedresearchprojectin1969isnowmoreprivatethanpublic,andapotentsocial,cultural,technological,andeconomicforce.
11.
1EffectsonBusinessModelsTheinternetecosystemhasbroughtnewbusinessmodelstothefore,allofwhicharerepresented,tovaryingextents,inthisreport.
Onecouldsaythatifthereisadefiningcharacteristicofdigital,networkedbusinessmodelsitisthattheydefytheconventionaleconomicwisdomsoferasgoneby.
Whatelseexplainsthemulti-billiondollarbusiness-esbuiltonfoundationssuchashandingthekeystoyourhometocompletestrangersorprovidingservicesfreeofchargetoallbutthreepercentofyourcustomerbaseNewinternet-reliantbusinessmodelsinclude:Contentdiscoveryplatformsthatgivedigitalpublishersaccesstonewaudiencesatscale,suchasRefinery29,BuzzFeed,andEliteDaily.
TheplatformeconomicsthatunderliebusinessessuchasUberandAirbnb,andthenewrangeofchoicestheymakeavailabletoconsumers.
Chapter11:SocietalBenefitsoftheInternetEcosystem112Freemiumbusinessmodels,buoyedbythetinycostsofservingincrementaldigitalcustomers,enablethebuildingofprofitablebusinesseswithjustasingledigitper-centageofcustomerspaying.
Flexible,self-scheduled,independentwork,whetherintheformofdeliveryservicessuchasInstacartandTaskRabbitoronplatformsthroughwhichcontractemploymentcanbeobtained,e.
g.
Behance,Upwork,andHourlyNerd.
Theblockchain,apublic,distributedledgerofdigitaltransactionsthatallowsinstantverificationstotakeplacewithoutacentralizedauthority;alreadyinusebyfor-ward-thinkingartistsanddozensoftopfinancialinstitutions.
IaaS(InfrastructureasaService):Capabilitiesthatcanbeaccessedviathecloudenablestartupswithmodestbudgetstouseenterpriseleveltechnologyatafractionofearliercosts.
Thesenewbusinessmodelsmakepossibleavarietyofenterpriseactivitythatcouldnotexistwithouttheinternet.
Asamplingincludes:SellersoneBay,Etsy,Amazon,andCraigslistenumeratedintheOn-DemandEcono-myworkforcesectionofthisreport,whomayearnsupplementalincomeorfulltimeincomebycreatingand/orsellingitemsonline.
Themodularizationofservices,asseenwithonlinefirmssuchasTalkspace.
com.
BasedinNewYorkCity,Talkspacehasover1,000licensedtherapists,availableviatextandchatapps,sothatdealingwithpersonalproblemsisconvenient,inthemoment,andmoreaffordablethanface-to-facetherapy(e.
g.
$32perweekforunlimitedtherapyviamessaging,billedmonthly120).
ThePennyHoarder.
com:Whatbeganasapersonalfinanceblogwrittenbyacol-legedropoutgrewintoamediacompanywith$20millioninrevenueandoverfivemillionactivesubscribers.
Thecombinationoffreecoursesand'nanodegrees',orhighlyfocusedcertifications,offeredbyonlinelearningplatformUdacity,madepossiblethroughpartnershipswithtechcompaniesrequiringworkerswithskillstaughtinthecoursestheysponsor.
11.
2MatchmakingforSocietalGoodMatchmakingisanotherhallmarkofinternet-enabledsystems,abletoreduceandsome-timeseveneliminatetheinefficienciesandfrictionsoftransactionsinthephysicalworld.
Notcoincidentally,oneofthefirstbusinesscategoriestoflourishonlinewasdating;tra-ditionalmatchmakingaugmentedbydigital,networkedtechnologies.
Todaythematch-120https://www.
talkspace.
com/online-therapy/faqs/-t4113makingcapabilitiesoftheinternetextendbeyonddatingtothelabormarket,personalsafety,andsocialinclusion.
Examplesinclude:TechHire:WhileprivateeducationalinstitutionspromotecoursesandtraininginSTEM(Science,Technology,Engineering,andMath),TechHireisaU.
S.
govern-ment-supportedinitiativethataimstoaddressthegapbetweentheclosetosixmillionjobopeningsinthetechsectorandthetensofmillionsofAmericansseekingemployment.
AcrossallstatesTechHireservesasanetworkthroughwhichemployerswithhiringneeds,traininginstitutions,andjobseekersconnect.
Manyofthejobseekersarenotfromtraditionaltechbackgrounds,andtheprogramencouragesem-ployerstohire"basedonmastery,nothistory…[and]adaptcurriculatorespondtoemployerdemandsignalswithprogramsthatfocusoncompetenciesandjobplace-ment.
"121SitWithUs:Someinternet-enabledservicesaimtosolvethebigproblems,suchasTechHire,whileothersarecontenttotakeonthesmallerproblems,suchastheun-happinessbroughtonbyeatinglunchaloneinthehighschoolcafeteria.
Developedby16-year-oldNatalieHamptonofShermanOaks,California,SitWithUsisanappthatmakesfindinglunchtablebuddiesdiscreetandeasy.
Circleof6:Afreesafetytooloriginallydesignedtohelpcollegestudentsleveragetheirsocialnetworksforprotection.
Forexample,twotapsinsidetheappletsyournetworkknowthatyouwanttobeinterrupted,therebybeingprovidedwithawaytoexitapotentiallydangeroussituation,e.
g.
atabarorclub.
TheappisnowwidelyavailableforiOSandAndroidandisbeingusedbeyondstudentsandcampuses.
11.
3CommunitiesofInterest:AdhocOrganizingTheabilitytobringtogetherlike-mindedindividuals,whetherwithinthesamegeographyordispersedacrossgreatdistances,isanotherdefiningtraitofinternet-reliantbusiness-es.
Theresultingsocietalgoodaccruesbenefitstocommunitiesintheformofinnovation,publicpolicy,andevenuniquemethodsoffinancing.
Projectsandinitiativesthattakeadvantageofadhocorganizinginclude:Reinvent.
net:Anopencommunityofinnovativemindsinterestedinconvening"…amixofsmart,knowledgeable,innovativepeoplefromawiderangeoffieldstoworkonsolvingthebigchallengesofourtime.
"122Thosechallengesincludemanufactur-ing,health,politicalsystems,foodandwatersupplies,andmore.
121http://techhire.
org/about/114Peers.
org:Thisorganizationaimstomaketheeconomyofon-demandworkandservicespracticalandsustainable.
Initiativesincludesystemsthatmakefindingandmanagingflexibleworkeasier,andfacilitatingportablebenefitsforindependentworkers.
CivicMedia:Thisprogramseekstoenhanceparticipationinciviclife,onlineandoffline,organizing,communicating,andexecutingplansandprojectsthatenhancetheexperienceoflifeincitiesandaddanewdimensiontocitizenry.
Samplecivicmediaprojectsinclude:ActionPath,alocation-basedplatformthatsurveyscitizensforfeedbackoncom-munityprojects.
123CommunityPlanIt,anonlinegamethatplacesplayersintherolesofadiversegroupofstakeholdersandchallengesthemtobuildthenecessarytrustandskillstocompletetime-limitedcivicplanningmissions.
124GetCalFresh,anappthatsupportsthetwomillionCalifornianseligibleforfoodassistancebystreamliningtheapplicationprocess,providingefficientclientser-vices,andtreatingthoseinneedwithdignity.
125Indiegogo'spartnershipwithMicroVentures126:Crowdfundingisnothingnew,buttheabilityforanyonetoholdanequitystakeinanewandexcitingproductorcompanyis.
Ratherthanjustreceivingarewardlevelperkand/ortheproductifacampaignissuccessful,thisprogramletscrowdfundersbenefitfromsupportinganinnovationearly.
Indiegogohasraisedover$1billionforover200,000projectsrangingfromappsandgamestophysicalproducts.
Whilenotexhaustive,thislistprovidesaviewintoaninternet-enabledworldofcom-munityandopportunitythatextendsbeyondtraditionaldefinitionsofcollaborationandevenofbusiness.
Weexpectsuchinitiativestocontinuetodevelopasdevicesbecomecheaper,smaller,andmorepowerful,andconnectivityisbuiltintowearabledevicesandhouseholdobjects.
122http://reinvent.
net/about123https://civic.
mit.
edu/action-path-0124https://elab.
emerson.
edu/projects/civic-media/community-planit125https://www.
codeforamerica.
org/products/getcalfresh126https://equity.
indiegogo.
com/115Theconclusionsofthisreportareorganizedaroundthetaskssetoutintheprojectassignment:1.
Definetheadvertising-supportedinternetecosystem.
2.
Determinetheemploymentandvalueofeconomicactivityenabledbythisecosystem.
3.
Determineitsgeographicdispersion.
4.
Definethecompaniesassociatedwiththead-supportedinternetecosystem.
5.
Determinethecontributionofadvertisingtotheecosystem.
12.
1DefiningtheAdvertising-SupportedInternetEcosystemAsdescribedinChapter1,theinternetecosystemisaninternet-dependentmarket-mak-ingandbusiness-enablingsystemwithinthebroadereconomy,definedbyactivitiesthatrelyontheinternettopromoteexchangesofproducts,services,andinformation.
Itisstrikingtofindthattheecosystemsodefinedhasgrownatafasterrateinthepastfouryearsthaninthepreviousfouryears.
AsChapter8reports,thegrossnationalproducttheU.
S.
hasaccelerateditsdependenceontheinternet,reflectingthefactthatitisincreasinglyadata-driveneconomy.
Theecosystem'sstructureisbroadlyunchangedfromearlierstudies,withfirmsconcen-tratedinfourhorizontallayers.
However,aswediscussinChapter7,anumericallysmallbuteconomicallylargegroupoffirmshasbeguntointegrateverticallyacrossthehorizontallayers.
Thesefirmscompetewitheachothertoprovideintegratedsolutionstothemarketingchallengesofbrandmarketers,butcompetealsowithadhocalliancesoffirmsinthehorizontallayers.
12.
2TheEmploymentandValueofEconomicActivityEnabledbythisEcosystemTheecosystememploys4.
1millionpeopledirectly,and6.
3millionpeopleindirectly,aslaidoutinChapter8.
Itcontributes$1,121billiontoU.
S.
GDP.
Bothemploymentandeconomicvaluehavegrownatanacceleratingratesinceourfirststudyin2008.
Threesub-sectorsoftheecosystemstandoutasdriversofthisacceleration:digitalcom-merce,cloud-baseddeliveryofsoftwareandservices,andstreamingconsumptionofentertainment,buthardlyanyelementoftheeconomyhasbeenuntouched.
Thepaceofdata-drivenentrepreneurshipinareassuchasIoT,bots,financialtechnology(FinTech),marketingtechnology(MarTech),andinpublishing,assitesofmediaconsumptionChapter12:Conclusions116becomeuncoupledfromthesitesofproduction,allsuggestthatacceleratedgrowthwillcontinuefortheforeseeablefuture.
Whenanalyzedbyhorizontalsector,jobgrowthwasfastintheconsumer-facinglayer,amonghouseholdnameslikeFacebook,YouTube,UberandAirbnb.
However,itwasfasttoointhelessglamorousSoftInfrastructureandConsumerServicesSupportlayersthatunderpinthehigh-profilebrandnameconsumerservices.
Consumerscouldgetthebenefitsoftheinternetatlowercost,oreven,forserviceslikeYelp,Facebook,Twitterandmanyothers,forfree,becauseentrepreneurswerebuildingoutsupportservicestomakethemleanerandmoreprofitable.
Itmayevenbe(thoughourreportdoesnotproveit)thatgrowthinthesesupportlayersmadeitunnecessarytogrowemploymentinconsumer-facingservicesasfastastheirrevenuesgrew,ifitmadetheseservicesmoreefficient.
Whatisclearisthatdatadrivesinternetinnovation,andarichsub-ecologyofconsum-ersupportservicesgrowsuptomakesenseofthedata.
Inmattersofregulationitisasimportanttoconsiderimpactonthesupportservicesasontherestofthestructure.
12.
3TheGeographicDispersionoftheAd-supportedInternetEcosystemByrelyingontheU.
S.
CensusBureau'sallocationofinternet-intensiveNAICScodeemploymenttothecountiesoftheUnitedStates,andmappingcountiestocongressionaldistricts,wefoundinChapter10thattherewasinternet-relatedemploymentineverycongressionaldistrictoftheU.
S.
Thedispersionoftheworkforcewasbroad,with116congressionaldistrictsaccountingforhalfoftheemployment,and320accountingfortheotherhalf.
Certainly,thepopularconceptionthatinternetworkisconcentratedinSiliconValleyisnotfalse.
FivecongressionaldistrictsontheSouthSanFranciscoPeninsulaemploy174,000ofthepeopleidentifiedinthecensusdata.
Andthereareotherconcentra-tions,suchasthreedistrictsinManhattanthattogetheraccountfor106,000people,theSeattle/Tacomacluster,theAustin/SanAntonioclusterinTexas,andBoston'sRoute128corridor.
Butitwouldbewrongtooverstatethedegreeofconcentration.
Some86percentofemploymentisoutsideofthe14mostinternet-intensivecongressionaldistricts,spreadthroughouttheentirenation.
12.
4TheCompaniesAssociatedwiththeAd-supportedInternetEcosystemIncontrasttomethodsthatrelyexclusivelyonthesurveysoftheU.
S.
DepartmentofCommerce,whichrespecttheanonymityofrespondentestablishments,ourmethod117enumeratesbynamethe400orsolargestfirmsintheinternetecosystem.
Anditpointstomanysmallercompaniesthatcontributetothevibrancyoftheecosystem.
AsnotedinChapter8,ourenumerationoflargefirmsproduced2.
3million,or56percent,ofthe4.
1millionpeopledirectlyemployedintheinternetecosystem.
Theremainderrangedfrommid-sizedandsmallfirmstoself-employedindividuals.
12.
5TheContributionofAdvertisingtotheEcosystem.
Inthepre-internetworld'advertising'referredtopaymentstomedia.
Itdidnotcoveradvertisingonso-called'owned'mediasuchasdisplaysonthesidesofafirm'strucksandbuildings,nordiditcoverdirectmail,catalogretailing,ortelemarketing.
Ifweap-plythisnarrowdefinitionof'advertising'aspaymentstomediaforservicesthatincludesearch,display,classified,mobile,leadgenerationandemail,thenadvertisingcontrib-uted$65billiontotheecosystem.
Butadvertising,whendefinedmorebroadlyasmarket-makingandbusiness-enabling,isthewholeoftheecosystem.
Itseffects,includingthoseofownedandearnedmedia,arereflectedinits$1,121billioncontributiontoGDP.
Paymentstothird-partymediamea-surelittleofwhattheinternetdoestomakethemarketsthatcreatetheeconomy.
Theinternet,insum,servesmanymarket-makingpurposesbesidesmediaadvertisingnarrowlydefined.
Websitescanserveasstorefronts,point-of-purchasestimuli,astoolsforconductingresearchonlineforofflinepurchase,andtotransactonlinebasedonresearchoffline.
Websitescanaggregateconsumerreviews.
Consumerscanseeprod-uctspromotedandbuytheminthesamevisitiftheychoose.
Theycandownloaddigitalproductsandconsumethemonline.
Theycansharenewsabouttheirpurchasesandopinionsandreviewproductsandservicesonsocialmedia.
12.
6ConclusionThereislittlenewsintheobservationthattheinternethasimpactedcontemporarylifelikenoothertechnicalproductinlivingmemory.
Withthisreportanditstwoprede-cessorswehavetriedtogofurtherandquantifyitsimpactatthreepointsintime.
Wehopethatbydeepeningtheunderstandingoftheinternet'sroleintheeconomyanditsstructure,wehaveprovideddataandanalysisthatwilladvancethediscussionofpolicychoicesastheadvertising-supportedinterneteconomycontinuestoextenditsreachandbecomealargerandmoreintegratedpartoftheU.
S.
economy.
118JohnDeightonisaBakerFoundationProfessorofBusinessAdministrationatHarvardBusinessSchool.
Thefocusofhisresearchisondigitalanddatabasemarketing.
TheresearchhasbeenpublishedinavarietyofjournalsincludingJournalofConsumerRe-search,JournalofMarketingResearch,JournalofMarketing,JournalofInteractiveMar-keting,OrganizationalBehaviorandHumanDecisionProcesses,andHarvardBusinessReview.
Ithasreceivedanumberofcommendations,includingtheAmericanMarketingAssociation'sBestArticleAwardforanarticleinJournalofMarketingandanhonorablementionfromJournalofInteractiveMarketing.
HewasfoundingeditorofJournalofInteractiveMarketingandatwo-termeditorofJournalofConsumerResearch.
HehasbeenavisitingscholarattheUniversityofTokyo,DukeUniversity'sFuquaSchoolofBusiness,CambridgeUniversity'sJudgeSchoolofBusiness,andOxfordUniversity'sSaidBusinessSchool.
HeisapastExecutiveDirectoroftheMarketingScienceInstitute,amemberoftheChairman'sAdvisoryCouncilofMarketingEdge,andaDirectoroftheBerkmanKleinCenterforInternetandSocietyatHarvardUniversity.
PriortojoiningHBS,hewasonthefacultiesoftheUniversityofChicagoandDartmouthCollege'sTuckSchoolofBusiness.
HehasaPh.
D.
inMarketingfromtheWhartonSchool,UniversityofPennsylvania,anMBAfromtheUniversityofCapeTown,andaB.
Sc.
inChemicalEngineeringfromtheUniversityofNatal.
Hisappliedresearchin-cludesconsultingwithanumberofU.
S.
andinternationalcorporations.
LeoraKornfeldisanAdjunctFacultyMemberatYorkUniversity'sSchulichSchoolofBusinessandhasservedasco-authororPrincipalInvestigatorforseveralstudiesfocusedontheinterneteconomyinNorthAmerica.
Otherpublicationsinclude20HarvardBusi-nessSchoolcasesintheareaofdigitalmarketingstrategyandtheJournalofInteractiveMarketingHonorableMentionforBestPaper.
SheholdsanM.
A.
inMedia&Communi-cationsfromtheUniversityofLondon,GoldsmithsCollege.
MarlonGerraisaCharteredFinancialAnalystwithaspecialtyincompanyanalysis,industryresearch,andadvanceddataanalysis.
HeholdsanM.
A.
instatisticsfromHar-vardUniversityandanM.
Sc.
inFinancialEngineeringfromtheUniversityofMichigan.
AbouttheAuthors