ModelingCustomerRelationshipsAFlexible,IntegratedArchitectureEnablesCustomer-CentricMarketings1AbstractThiswhitepaperdrawsuponthelessonslearnedbySequentComputerSystemsinimplementinglarge-scaletechnologyplatformstosupportCustomerRelationshipManagement(CRM)strategieswithinmajororganizationsworldwide.Itisassumedthatthereaderisfamiliarwiththestrategicdirectionofmostcustomer-focusedorganizationsandunderstandsthecyclicalandrepeatablenatureofatechnology-drivenmarketingstrategy.Interestedreadersarelikelytobeconcernedwithhowtheymightmodelcustomerrelationshipsinawaywhichwillsupportthetransitionfrompresentproduct-focusedviewsofmarketingintelligencetomoreuseful(andprof-itable)customer-focusedviews.Thispaperprovidesthedataarchitectormodelerwithagenerictemplateformodelingcustomerdata.Thisapproachisnotproductortechnologyspecificbutdoesprovideaflexibledataarchitectureforintegratingthevarioustechnologycomponentsthatusedatatodrivemarketing.Thispaperalsohighlightscommonlyfacedproblemsthatoccurwhenmodelingcustomerdata.Editor:DavidPuckeySequentComputerSystemsUKProfessionalServicesIntroductionThestrategicimportanceofmanagingcustomerrelationshipsbothdrivesandisdrivenbytechnology.Inpar-ticular,thisappliestodataandtheincreasinglysophisticatedandusefulwaysinwhichdataisusedtomodelrelationshipsandtodrivecontactstrategies.AtthecoreofanytechnologyenablerforCRMisthecustomerdatabase.Thecustomerdatabaserep-resentsthedatahubthatintegratesthevariousstatisticalmodeling,cam-paignmanagement,contacthistoryandresponsetrackingcomponentsofthemarketingcampaignlifecycle.Thisistruewhetherthedatabaseisusedfortheexecutionofmarketingstrategies(e.g.,generatesmailinglists),orwhetheritexistspurelyasananalysisenginethatpassescontactstrategiesandinformationtoaseparatecustomerinteractionplatformforexecution(e.g.,customercallcenters).Thetechnologylayeranditsintegrationwithemergingbusinessprocessesisthereforekeytothesuccessfulimple-mentationofadata-drivenCustomerRelationshipManagementstrategy.Thispaperdescribes,inagenericway,anapproachthatSequenthassuccessfullyusedtoensurethatthecoredatastruc-turesintheCRMtechnologylayersupporttheintegrationofthevariouscomponentsofthemarketingprocessandreducethetimerequiredtodesignandexecuteacampaign.Thisapproachenablesthecreationofacompletemodelofcustomerrelationshipsovertime.TheEvolutionoftheCustomerDatabaseCurrentapproachestothedesignofthecustomerdatabasefallbroadlyintotwocamps.Thefirst—theflatearthviewoftheworld—hailsfromtheglorydaysoftargetmarketinginthelate1980s.Thisapproach,whichispopularwithlistprovidersandbureauoperations,utilizestheconceptofthecustomerfileorlist.Suchalisttendstoofferacurrentsnapshotofthecustomerorprospectbaseandisoftentheproductofmuchtortuouscleansing,de-duplicationandpoint-in-timesegmentation.Thisapproachmakesitdifficulttoanalyzetheups-and-downsofanorganization’srelationshipwithacustomerovertimeduetoitscurrentsnapshotviewofthecustomerandprospectbase.Further,ittypicallylacksthetransaction-leveldetailandpromotionalhistoryneededtomodelcustomerbehavior.1Figure1:CustomerdatabasesevolvewithintegrationoftechnologyandbusinessprocessesSource:RaabandAssociatesncontacthorizonnoutputnsystemsnexecutionndepartmentsndatatypesnupdateintervalnreactiontimengoalCampaignManagementnone-shotnoffernmail/phonenmanualnmarketingnpurchasesnmonthlynbillingcyclenreducewasteCustomerRelationshipManagementnsequenceninformationntouchpointnschedulednfront-officencontactsndailyntransactionnaddrevenueEvolutionofCustomerDatabase2ThesecondapproachhasevolvedfromthedatawarehousingmovementandSequent’sexperienceinhelpinghundredsoforganizationsdesignandimplementdatawarehouses.Sequenthasdevelopedamaturemethodologyfordeliveringrapidbusinessbenefitbyintegratingsophisticatedanalyticaltoolswithsubject-orientedandtime-consistentcentraldatabases.Suchsystemstypicallyconcentrateonthedeliveryofbusinessintelligenceandaregenerallynotdesignedtoplugdirectlyintoanorganization’sday-to-dayoperations.However,themodelingtechniquesemployedbySequentforthedeliveryofsuccessfuldatawarehousingprojectsrepresentaradicalshiftinemphasisfrombothflatearthviewsofdataandthemicroscopicviewsofdatausedinonlinetransactionprocessing(OLTP)systems.Sequent’sdimensionalviewofdataprovidestheoptimumcombinationofanalysisoffactsovertimeandhighsystemperformancewhendealingwithlargedatavolumes.Sequent’sapproachtosuccessfullydeliveringlarge-scaletechnologyplatformstosupportCRMstrategiesusesthebestattributesfrombothofthepreviousapproaches.Thisapproachincorporatesthemaximumdegreeofanalyticalflexibilityforthemarketerandmarketinganalystwiththeefficientscoring,segmentationandextractionofdatatoexecutemarketingcampaignsorcontactstrategies.Italsoplacesthecustomerorprospectatthecenterofthemodelandseekstomodelallfacetsofarelationshipwiththatcustomerovertheknownlifetimeoftherelationship.Sequent’sapproachtothedesignofcustomerdatabasesisnotlistbasedandisnotdesignedtosimplysupportad-hoc,point-in-timemarketingsolutions.Rather,theobjectiveistogivethemarketertrueinsightintothevari-abilityofhisrelationshipwith