Marketing Experiment Assignment - Christian van 't Hof
-
Upload
christian-van-t-hof -
Category
Documents
-
view
111 -
download
0
Transcript of Marketing Experiment Assignment - Christian van 't Hof
Student:Christianvan’tHof 11thofJanuary2016
MarketingExperimentAssignment
1IntroductionThisreportisaboutamarketingexperimentforVikingBedsofSweden.ThisSwedishboxspringbrandhasjustbeenintroducedontheDutchmarketintheupper-middleandhighersegment.ThemarketingwillbeaTVadvertisingcampaignfortheirboxspringtype‘RoyalViking’.Sinceboxspringsaregainingmoremarketshare,thesehighqualitybedsarebeingthe‘realdeal’.Theexperimentthatisgoingtorunwillbeabefore-afterdesignexperiment.Sincetherehasn’tbeenanybrandexposureyet,abasicexperimentwon’tbeworkingbecausethecitywiththeTVadswillalwaysbesellingmore.Moreover,afullfactorialexperimentwon’tbeworkingeither,becauseit’sabout1productwiththesamepricingforbothcities.Inthiscase,thebefore-afterdesignexperimentwillbeworkingbyshowingtwodifferentads.
2ExperimentDesignThevariablesinthisexperimentwillbeindependentaswellasdependent.ThedependablevariablewillbetheamountofunitssoldandtheindependentvariablewillbetheTVadvertisements.ThisexperimentwillruninthecitiesLarenandNaardenintheNetherlands.Thesecitiesarebothinthesamegeographicalarea,sharealmostthesameamountofpopulationandhavebothonaverageabout€50,000incomeperhousehold1,whichisalmostthehighestintheNetherlands.BothLarenandNaardenhaveabove10,000inhabitants,whichisenoughforstatisticalsignificance.TheentirecitieswillseethatTVads,sothismakesthisavalidexperiment.ThefirstroundofTVadswillrunfromMarchtillAprilandthesecondroundofTVadswillrunfromMaytillJune.Inthefirstround,bothcitieswillseeTVAd1.InthesecondroundLarenwillseeTVAd1andNaardenwillseeTVAd2.Theliftinsaleswillbecalculatedfromthedifferencesbetweenthetwoperiods.Intermsofcausalitythreeruleswillatleastbeapplicable,theotheroneisinthiscasenotapplicable.Thetwoapplicablerulesarethefirst,secondandthethirdone:Changeinthemarketingmixproduceschangeinsales,nosalesincreasewhenthere’snochangeinthemarketingmix(Round2,Laren)andthetimesequence.Inthishigh-endmarket,competitorswillprobablyrespond,sothatwillbetheexternalfactors.
1 Centraal Bureau voor Statistiek. (2015, December 8). StatLine - Centraal Bureau voor Statistiek. Retrieved
Student:Christianvan’tHof 11thofJanuary2016
3AnticipatedIssues3.1-2IssuestobeconsideredInthisexperimenttherecouldbeissues.IntheNetherlandspeoplegetaroundMayandJune,theirholidayfeefromtheirjobs,thiscouldincreasetheirbudgetinthosemonths,whichcouldbeanexternalfactorthatcouldaffectthenetlift.AnotherissuecouldbeourcompetitorCarpeDiem.CarpeDiemisalsofromSwedenandhasalreadyenteredthemarket.Iftheylaunchanewcampaign,thanthiswoulddefinitelyaffectthenetliftaswell.
3.2-2ThingstheexperimentdemonstratesThisexperimentcouldpointoutadifferenceinoutcomeofbothTVAds.ItcouldindicatewhichTVAdwouldworkbetter.Ontheotherhand,aresultofthisexperimentcouldalsoindicatewhichcitywouldbebuyingmoreproductsfromourcompanycomparedtotheother.Iftheoutcomeofbothcitiesshowsverydifferentnumbers,themostinterestingcitycouldbeappointed.
4ExperimentAdaptionInthis2ndversionVikingisgoingtoshowtheirTVAdsontheInternet.Thiscampaignwillspaninthesamemonthsinthesecondyear.Inthiscase,afullfactorialexperimentwillassistinprovidingmorespecificmetrics,whichcanbecontrolledmoreprecise.Inthiscasethevideoadswillbeanindependentvariableaswellasgender.AccordingtometricsSweden,malestendtobuytheseboxspringsoverfemales.Thedependentvariablewillstillbetheamountofunitssold.Thismorespecificexperimentwillshowmorerelevantinformationonthegenderofourcustomers,whichwecanusetoimprovethetargetingofcustomersbymorespecifiedvideosthatwillbemorefocussedonthedifferencesingender.Theresultsfromthisexperimentwillbeeasiertomeasureandtoanalyse,whichsavesthecompanymoneyandgivesthecompanybetterresults.Ontheotherhand,justlikeTVads,itonlymeasuresthepeoplethatusetheInternetontheircomputerorwatchTV.