基于客户终生价值的精确细分研究及其在电信行业的实证研究

上海交通大学硕士学位论文基于客户终生价值的精确细分研究及其在电信行业的实证研究姓名：杨文杰申请学位级别：硕士专业：企业管理指导教师：黄沛20080101I3GCLV(CustomerLifetimeValue)IIACUSTOMERSEGMENTATIONMODELRESEARCHBASEDONCUSTOMERLIFETIMEVALUE:ACASESTUDYONTHEWIRELESSTELECOMINDUSTRYABSTRACTOwingtothecomingof3G,intensecompetitionwouldoccurinthedomestictelecommunicationmarket.Thethingthatreallymattersinthecompetitionistheownershipofhigh-qualitycustomersandcustomersegmentationistheprerequisiteandfoundation.Atpresent,thecurrentsegmentationmethodissimplyanexperienceandstatisticsbasedsystem,whichdoesnotinvolvethefuturevalueofthecustomerandhislifecycle.Asaresult,thesemethodsarecouldnotidentifytherealhigh-qualitycustomersandthedifferentiatedmarketingstrategywouldnotbeaseffectiveaspredicted.Thus,toidentifytherealhigh-qualitycustomers,itiscrucialforatelecommunicationcarriertoestablishapreciseandsmartcustomersegmentationmodelbasedontheCLV(customerlifevalue)theoryaswellasthedataminingtechnology.Basedonthetheoriesofcustomersegmentation,customersegmentationmethodsandcustomerlifetimevalue,thepaperstudiescustomersegmentationmodelontheChinesetelecommunicationindustryfromthecustomerlifetimevalueperspective.Firstly,thepaperimprovescustomerlifetimevaluemodeltoputforwardanewmodel.Thepaperchangesthepotentialvaluemodeltomakeiteasytocalculate;thepaperalsointroducesthecustomerloyaltyandcustomercrediteffecttotheCLVmodel.What’smore,dataminingtechnologyisusedwhenbuildingthemodel.It’sdeeplybelievedthatthemodifiedmodelcanreflectthestateofthecustomers’valuemoreobjectivelyandrealizeclassificationofcustomerswithdifferentvalues.Secondly,basedonthemodifiedCLVmodel,thepaperputforwardathree-dimension(customercurrentvalue,customerpotentialvalue,customerrelationshipvalue)customerclassificationandmadeanin-depthresearchinthedifferentiatedmarketingstrategybasedonthecustomerlifetimevaluesegmentation.Atlast,thepaperelaboratestheoreticallythestrategicimportanceontheapplicationofconsumersegmentationintelecommunicationbusinessbasedoncustomerlifetimevalue.Toprovethemodelandthetheory,thepaperappliestheCLVsegmentationintoaChinaUnicomcarrierandachievesaverygoodfeedback.Byusingtheabovemodelstoclassifycustomers,leadersofthecorporatecanidentifyvaluablecustomersandrecognizecustomer’sbehaviormoreeffectively.Theycantakedifferentmarketingstrategiesaccordingtodifferentvaluablecustomers,andputlimitedresourcetothemostvaluablecustomers,soastoenhancethecompetitivenessofthecorporateandbringlong-terminterests.KEYWORDS:CustomerLifetimeValue,CustomerSegmentation,DataMing,DifferentiatedMarketingStrategy-1-1.1,,WTO,,,,,,,,3G-2-1.21.2.11a)SMS20%80%b)21.31.3.11.(WendellR.Smith)19561-3-:YoramWind()()/:2a),b)LazerWellsTigertAIOPlummerc)(1)RFMHughesRFMR-4-;F;MRFM(2)2RFMMarcusRFM5RFMR1-1RFMMarcus1-1RFMRFM;;;(3)LRFM(2003)LRFM6LRFMRFM(L)d)Griffin71-2:-5-1-2GremlerBrown()()()8()()()e)BostonConsultingGroup1-36012341-3f)Haley-6-Reichheld199341.3.2(Value-basedSegmentation)1010%20%80%1010%10%1%2(1)(2)3.1-4.()-7-1-4(Up-buying)(Cross-buying)41-4():1%():4%():15%():80%()-8-2.1:()VRCV=R-C9(CurrentValue):(PotentialValue):2-12-1BarbaraJackson(1985)Roberts&Berger(1989)Dwyer(1989)Bitran&Mondschein(1996)Berger&Nasr(1998)Reichheld(1996)Gupta&Lehmann(2003):(RemainingLife);(LifeCycle)Courtheoux(1995)10.(CustomerLongtimeValue).-9-2-12-1Gupta&Lehmann(2003)2.2:(1)AngusJenkison(1995)11n-iii1CLV=C(1d)=+∑(2.1)iCind(2)Wayland(1997)CLV-10-12CLV:nniiiiii1i1CLV[GCPr(i)(1d)(RD)(1d)A]−−===××+−+×+−∑∑(2.2)iGCiPr(i)iAiR,iDi(3)(2001)1314:TPaulD.Berger&NadaI.Nasr(1998):12111tt211122322222(t)(t)htv,0tt(t)(t)(t)[N(1e)],ttt(t)(t)(t)h(tt),ttTππππππππ−+==+≤≤==+−≤==−−≤(2.3)1h2hNv.vv0-1t,;1t-2t2t(4)Jackson(1985)(Lost-for-good)(Always-a-share)15::CLV:31212ttt123t=0t=tt=tCLV=()()(1)()()(1)()()(1)iiitrtdtrtdtrtdπππ−−−××++××++××+∑∑∑(2.4)CLV31212ttt1230ttCLV=v+()()(1)()()(1)()()(1)iiitrtddttrtddttrtddtπππ−−−××++××++××+∫∫∫(2.5)()rtCLV:-11-31212ttt123t=0t=tt=tCLV=()Pr()(1)()Pr()(1)()Pr()(1)iiittdttdttdπππ−−−××++××++××+∑∑∑(2.6)CLV31212ttt1230ttCLV=v+()Pr()(1)()Pr()(1)()Pr()(1)iiittddtttddtttddtπππ−−−××++××++××+∫∫∫(2.7)Pr()t(5)HyunseokHwang,TaesooJung,EuihoSuhCLV16,CLV(Verhoef,Donkers,2001)CLV()i01()()CLV()(1)(1)iiiiiiiiiNNEifiiNtpitNttNtBttddππ+−−==++=+++∑∑(2.8)()1/()churnEiPi==1()churnPi−itiiNid()Eii()pitπti()fitπtIB(t)t()churnPiii(6)(2002)17CLV.::STaCLRC,(i=a,p,s,m):apsmapsmCLVCLRCLRCLRCLRCCCC=+++−−−−(2.9)iiCLR=STa××-12-CLVCLV(7)(2002)18n:nijijj1M[P],P1===∑(2.10)ijijPiiijS:tFT-tittt=0i1CLV=[(1+r)(S)R]=∑∑(2.11):t()rT()tFtitStitRt(8)(2002)19a)(CustomerPurchasingValue,PV):PV=xxb)(PublicPraiseValue,PPV);-13-.:V=xxc)(CustomerInformationValue,IV)()d)(CustomerKnowledgeValue,KV)e)(CustomerTransactionValue,TV)2.3:(1)AngusJenkinson(1995)CLV:Reicheld(1996)Berger(1998)ReicheldCLVReichheld(1996)(2)RobertEWaylandPaulMCole(1996)-14-(Acquisition,Development&RetentionCost,ADR);(3):(4)CLVCLVCLVCLV(5)CLV()()()2.4:(1)()()-15-(2)CLVCLVRFM(3)Jain21-16-3.1HyunseokHwang,TaesooJung,EuihoSuhCLV(CustomerLifetimeValue)CLV(Verhoef,Donkers,2001)CLV()i01()()CLV()(1)(1)iiiiiiiiiNNEifiiNtpitNttNtBttddππ+−−==++=+++∑∑(3.1)()1/()ch