More Related Content Similar to Crm and techology Similar to Crm and techology (20) Crm and techology1. Technology Innovation and Implications for Customer Relationship Management
Author(s): Baohong Sun
Reviewed work(s):
Source: Marketing Science, Vol. 25, No. 6, 25th Anniversary Issue (Nov. - Dec., 2006), pp. 594-
597
Published by: INFORMS
Stable URL: http://www.jstor.org/stable/40057196 .
Accessed: 28/08/2012 07:17
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
INFORMS is collaborating with JSTOR to digitize, preserve and extend access to Marketing Science.
http://www.jstor.org
2. Technology Innovation and Implicationsfor
Customer RelationshipManagement
BaohongSun
TepperSchoolof Business,CarnegieMellonUniversity,5000ForbesAvenue,Pittsburgh,Pennsylvania15213,
bsun@andrew.cmu.edu
Marketing Science
Vol.25,No. 6, November-December2006,pp. 594-597
issn 0732-23991eissn 1526-548X1061250610594
infjffijH
doi 10.1287/mksc.l050.0165
©2006INFORMS
Invited Commentary
Keywords:customerrelationshipmanagement;servicechannel;CRMprograms;pricingof services;
developmentof customerdemand;cross-selling;communicationcampaign;adaptivelearning;dynamic
marketinginterventions
Existingresearchon servicesand relationshiptreats
customerservice as a majoroperatingvariable,and
focuseson measuringthe resultingcustomersatisfac-
tion, retention,duration,repeat purchases,word-of-
mouth (e.g., Bouldinget al. 1993).The recentarticle
by Rust and Chung (2006)gives an excellentreview
of existingmarketingmodels of serviceand customer
relationshipmanagement(CRM).
The rapid advancesin informationand communi-
cation technology provide greater opportunitiesfor
today's firms to establish,nurture,and sustain long-
term relationshipswith their customers than ever
before.The ultimate goal is to transformthese rela-
tionships into greaterprofitabilityby reducing cus-
tomer acquisitioncosts, increasingrepeatpurchases,
and charginghigher prices (Winer2001).First,CRM
is no longer the privilege of the service sector.Real-
izing the increasingimportanceof customerorienta-
tion, firms of all kinds of industries, ranging from
manufacturingto information,are exploringservice-
led growth as a promising means of differentiation
(Sawhney et al. 2004). Second, service is no longer
definedas a stand-alonemarketingdecisionaimed at
increasingcustomersatisfaction.Contemporaryprac-
tice of CRM has been integrated into every step
of the marketingprocess- handling productinquiry,
telemarketing,advertising,sales, transaction,service,
and survey. Third, current CRM has shifted from
staticrelationshipto dynamic"learningrelationship,"
from mass-marketingto customer-centricmarketing,
and from reactive service to proactive relationship
building.
The challenge faced by today's CRM practiceis:
How to learn about individualcustomersand act on
a firm'sknowledge of a customerfor the purpose of
growinga relationshipand improvinglong-termcus-
tomer value. The recentdevelopment of CRMtech-
nology calls for rigorous researchto better under-
standthenatureof thisemergingindustryandtohelp
design the best marketingprograms. As Rust and
Chung(2006)point out, researchon serviceand CRM
will inevitablygrow as partof mainstreammarketing
science.Given theirthoroughreview,I will focus on
discussingnew possibilitiesfor CRMand its implica-
tions for marketingmodels.
1. Increasing Service Channels and
Design of Integrated
Communication Structure
Since the 1990s,companieshave been racingto add
24-hourcall centers,directmail, email,fax, and web-
pages. Automatedservice channels,such as Internet
access,voice-recognitionphone systems,and transac-
tion kiosks, are given special emphasis to encourage
self-service.Thisraisesnew researchissues:
First, how do customers develop channel prefer-
ence?Recentworkin this area(e.g.,Ansariet al.2005,
Sullivanand Thomas2004)focuses on the retailpur-
chase (catalogueversus email).However,selectionof
a sales channelto make a purchasefrom the firm is
quite differentfromthe selectionof a communicated
channel. An understandingof channel habit forma-
tion and channelmigrationshould shed new light on
the best design of channelstructure.
Although invited commentariesare not formally peer-reviewed
and representthe opinion of the author,authorswere carefully
chosen based on theiroutstandingexpertisein the areasof their
respectivecommentaries.
594
3. Sun: InvitedComments:TechnologyInnovationand Implicationsfor CustomerRelationshipManagement
MarketingScience25(6),pp. 594-597,©2006INFORMS 595
Second,what is the role of each channel in every
stepofthecustomerdecisionprocess,suchasinforma-
tion search,purchase,transaction,and postpurchase
service? Researchhas shown both complementary
andcannibalizingrelationshipsamongchannels.How
to successfullyblend the various functionsof multi-
ple communicationchannelsto deliver great service
and slash costs?
Third,how can customersbe steeredto theirmost
preferredchannels?How self-sufficientcustomersbe
directed to self-learning channels? Given the high
cost of customer communication,it is importantto
learn the unobservablecustomercharacteristicsthat
translateinto heterogeneouschannel preferenceand
to determinethe optimalallocationof resourcesover
multiple channels to improve the effectiveness and
efficiencyof channelmix strategies.
2. Innovative CRM Programs
Customer relationshipsare actually built and sus-
tainedwith CRMprogramsfor which channelsserve
as a delivery mechanism. We observe more and
more novel CRM programs aiming at maintaining
long-termcustomerrelationshipssuch as loyaltypro-
grams, warranty programs, customization, rewards
programs,cross-selling campaigns, and community
building. How do customers respond to these pro-
grams?How canthe effectivenessof eachprogrambe
measured?How do these programshelp build long-
term customerrelationship?What is the best design
of these programs?Eachof these programsdemon-
strates unique propertiesthat have not been exten-
sively studied.
For example, a reward program is a promotion
strategy that encourages customers to accumulate
purchasesto obtaina reward.It is designed to sepa-
rate(current)purchasefrom(future)promotion.Cus-
tomers'currentpurchasingdecisions are determined
by futurerewardsthat they anticipate.Furthermore,
airlines, hotels, and car rentals advance the design
of rewardprogramsby allowing customersto choose
when and at what level to claim the reward.Accep-
tance of a reward becomes the customer's endoge-
nous decision. This endogenous decision is likely to
be drivenby the design, such as rewardamountand
distancebetween rewardlevels.
3. New Pricing Structure
As firms shift the purpose of offering service from
increasingcustomersatisfactionto growinga relation-
ship with customersand making profit,many inno-
vative pricing structuresemerge, such as pricing for
service upgrade and subscription pricing, advance
selling, and to introduce a long-term relationship
with customers.Analytical researchershave started
to examine the nature of these pricing tactics (e.g.,
Xie and Shugan 2001).Empiricalresearchis needed
to validate the nominal findings and measure their
effectiveness.
While Essegaieret al. (2002) and Danaher (2002)
provide good discussions on two-part pricing, re-
search is needed to investigate more sophisticated
designs of fixed fee and usage rate.Forexample,the
recentpricingstrategyadoptedby thefitnessindustry
and online DVD rental (Netflix) is no longer struc-
tured as price-per-unit.Instead,it is framedas pay-
ment forthe rightto gain accessto consumea certain
amount of services within a period of time. Prelimi-
naryempiricalevidencedemonstratesthatcustomers'
choiceof serviceplanis drivenby theirexpectedmax-
imumfutureconsumption.Theactualconsumptionis
much lower than the purchasedconsumptioncapac-
ity (Sunet al. 2005).
4. Managing Intertemporal
Evolvement of Customer
Demand and Cross-Selling
Researchhas shown thatcustomers'demandforvar-
ious products is governed by a latent and evolving
demand maturity,which develops with shift of cus-
tomerlife-stage,accumulationof productknowledge,
changeof financialresources,and consumptionexpe-
rience.A betterunderstandingof the developmentof
demand has importantimplicationsfor the design of
cross-sellingcampaignstrategy.
First,by gaining an in-depthunderstandingof the
evolving needs of each customer,the firmcanrecom-
mendthemostappropriateproductstobestmatchthe
unmet needs at differentstages of customerdemand
maturity.It helps the firm to forward-lookthe devel-
opmentof customers'futureneeds and providethem
with the right product,even before they realizethat
they need it.
Second, cross-sellingcampaigns have the indirect
effectof cultivatingcustomerneeds and arepartof a
multistagecustomereducationprogram.It is crucial
to take into accountthe indirecteducationeffectof a
cross-sellingcampaignto betterevaluateits effective-
ness.
Existingmarketingresearchhas focused on devel-
oping methodology to betterpredictpurchaseprob-
abilities for the product next to be purchased
(Kamakuraet al. 1991,Kamakuraet al.2002,Edwards
and Allenby 2003,Li et al. 2006).The goal is to find
thebestcustomersfora scheduledcampaign,with the
goal of increasingsales.See Liet al. (2005)fordiscus-
sions on modeling proactivecross-sellingcampaigns.
4. Sun: InvitedComments:TechnologyInnovationand Implicationsfor CustomerRelationshipManagement
596 MarketingScience25(6),pp. 594-597,©2006INFORMS
5. Adaptive Learning of Customer
Preference
Currenttechnology allows the firm to combine the
abilities to retrieve real-timecustomer information,
automatically analyze customer insights, respond
directly to customer requests and provide the cus-
tomer with a highly customized experience. This
offers the firm the opportunity to learn about cus-
tomer preferencein a real-time fashion. Sun et al.
(2006)term the use of accruinginformationto con-
tinuously update the firm's knowledge of customer
preferenceas adaptivelearning.As a result,the firm
becomessmarterwith experience.This learningrela-
tionshipwith customersallows the firmto bettercus-
tomize its CRMprograms.
To date, we have observed the booming of a soft-
ware industry specializing in data analysis and sta-
tistical learning tools that help firms to transform
their data storage system into a CRMdecision sup-
port system. Siebel,Epiphany,and Oraclehave filled
this CRMspace with software and hardwareprod-
uctsthatdo everythingautomaticallyfrompredicting
theirfuturemoves to recommendingrelatedproducts
to sendingdirectemailcommunications(Winer2001).
The emergenceof this data-miningtechnology calls
for a substantialamount of researchto develop sta-
tisticallearning rules for adaptive machine learning
and automatedimplementationof CRM.A firststep
is takenby the field of machinelearningwith appli-
cationsto databasemarketing(see a reviewby Witten
and Frank2005).
6. Integrationof Marketing and
Operations Management
Operationsmanagement(OM)models focus on min-
imizing operating cost and often ignore customer
reaction.However,in manypartsof the serviceindus-
try,operationaland marketingissues arehighly inter-
twined.Customerreaction(acquisition,retention,and
development)to a firm's cost controleffort,such as
product availabilityor on-time performance,needs
to be takeninto account.Effectivenessand efficiency
areequally importantin improvinglong-termprofit.
This opens up a vast agenda for multidisciplinary
research(Gans et al. 2003). Rust and Chung (2006)
point out thatOR-basedmodels in marketing-related
fields such as routing,supply chains, yield manage-
ment,and schedulinghave provided a basis for mar-
ketingmodels.
As an example,ever sincethe 1990s,the functionof
call centershas been transformedfrom simply deal-
ing with inquiriesto being a preferredand prevalent
channel for interactingwith their customers.Statis-
ticsshows that80%of a firm'scustomerinteractionis
throughcall centersand 92%of customersformtheir
opinion about a firm based on theirexperiencewith
call centers (PurdueUniversity2004).The operation
of call centersand the resultingcustomersatisfaction
and retentionshould be studied as crucialelements
of a firm's customerstrategy,akin to marketingand
loyalty programs.Sun and Li (2005) formulatecall
allocationdecisionsof a servicecenteras a CRMprob-
lem by integratingoperatingdecisions with market-
ing consequences.
7. Long-TermProfitImplications and
Dynamic Marketing Interventions
Existingliteratureon customerlifetime value analy-
sis calculatesnet present value of customers'future
profit(e.g., Kumaret al. 1997).Althoughfutureprof-
its are taken into account, many studies treat the
value as anothersegmentationvariableto guide tar-
geting strategies. Furthermore,as pointed by Rust
andChung(2006),mostof thecurrentmethodsignore
the endogenizationproblem that firm's intervention
changescustomer'sfuturepurchaseprobability.
Fora firm with the goal of maximizinglong-term
profit, CRM should be formulated as a stochastic
dynamiccontrolproblemunder demand uncertainty
with the firm as the decision maker which makes
dynamic marketing intervention decisions such as
pricing, channel strategy,or cross-sellingcampaign.
Thisframeworkallows thefirmto integratetheevolu-
tion of the latentdemandmaturity,the differentfunc-
tionsof the firm'smarketinginterventionsatdifferent
stagesof customerdemandmaturity,heterogeneityof
customerpreferences,thecostof acquisitionandlong-
term payoff with the goal of maximizing "lifetime"
profitof customers.
The marketinginterventionstrategyderived from
this frameworkis integrativebecauseit is a multistep,
multisegment,multichannelCRMprocessaboutwhen
to contactwhichcustomerwith whatproductor con-
tent using which communicationchannel (how).The
marketinginterventionis customer-centricbecauseit
follows each stage of customerdemandmaturityand
interveneswith the most appropriatemarketingtool.
Thesolutionis alsoproactivebecauseit allowsthefirm
to forecastthe effect of today's marketinginterven-
tion on futureprofitability.In addition,it can forego
short-termprofitin orderto preventthe loss of future
profit(e.g.,Lewis2005).Finally,theframeworkallows
the firm to be experimentaland learn aboutcustomer
preference.Forexample,the firmcan sampleby ran-
domly assigninga customerto a campaignchannelin
orderto learncustomerchannelpreference.
The formulationof CRMfrom the firm's perspec-
tive is summarized in Figure 1. Kamakuraet al.
(2005)give a good descriptionof existing methodol-
ogy on modelingcustomeracquisition,retention,and
development, which is representedby the expected
5. Sun: InvitedComments:TechnologyInnovationandImplicationsfor CustomerRelationshipManagement
MarketingScience25(6),pp. 594-597,©2006INFORMS 597
Figure1 FrameworkofCustomerRelationshipManagement
Notes.ZijkT:Marketinginterventionssuchas priceorcross-sellingcam-
paign.
8:Time-discountingvariable.
E[.]:Expectedprofit.
QiJT:Quantityofproduct/beingpurchasedbycustomer/at timet, includ-
ingno-purchase(QiJT= 0).
PRICEiJT:Pricepaidbycustomer/ forproduct/ attimer.
MCiJT:Marginalcostofprovidingproductj attimet forcustomer/.
INFOiJT:Informationavailabletothefirmaboutcustomer/ forproductj
attimer.
quantityE[QiJT IiJT,ZiJT]in Figure 1. Applicationof
CRM involves predictions of customer acquisition,
retention,and development probabilities,forecastof
future revenues, development of customer demand
and preference,firm learning and solution of mar-
keting interventionsin a dynamic setting. Thereare
many new managerial issues and methodological
challenges.Wehope thatthe summaryarticleby Rust
and Chung (2006)and this articlewill inspire more
comprehensiveand scientificstudies to address the
new challengesarisingfrom the growing practiceof
customerrelationshipmanagement.
References
Ansari,Asim,CarlMela,ScottA. Neslin. 2005.Customerchannel
migration.Workingpaper,TuckSchoolof Business,Dartmouth
College,Hanover,NH.
Boulding,W.,A. Kalra,R.Staelin,V.A. Zeithaml.1993.A dynamic
processmodelof servicequality:Fromexpectationsto behav-
ioralintentions./. MarketingRes.30 7-27.
DanaherJ. Peter.2002.Optimalpricingof new subscriptionser-
vices: Analysis of a marketexperiment.MarketingSci. 21(2)
119-138.
Edwards,Y.D., G. Allenby.2003.Multivariateanalysisof multiple
responsedata./. MarketingRes.40(3)321-334.
Essegaier,S., S. Gupta,Z. J. Zhang.2002.Pricingaccess services.
MarketingSci.21(2)119-138.
Gans,N., G. Koole,A. Mandelbaum.2003.Telephonecall centers:
Tutorial,reviewandresearchprospects(commissionedpaper).
ManufacturingServiceOper.Management(M&SOM)5(2)79-141.
Kamakura,Wagner,CarlMela,AsimAnsari,AnandBodapati,Pete
Fader,RaghuramIyengar,PrasadNaik,ScottNelsin,Boahong
Sun, PeterVerhoef,MichelWedel,Ron Wilcox.2005.Choice
models and customer relationshipmanagements- Summary
paper for the sixth choice symposium.MarketingLett.16(3)
279-291.
Kamakura,W.,S. Ramaswami,R.Srivastava.1991.Applyinglatent
traitanalysisin the evaluationof prospectsforcross-sellingof
financialservices.Internat.]. Res.Marketing8 329-349.
Kamakura,WagnerA., M. Wedel,F.de Rosa,J. A. Mazzon.2002.
Cross-sellingthroughdatabasemarketing:A mixed data fac-
toranalyzerfordataaugmentationand prediction.Internat.J.
MarketingRes.Forthcoming.
Kumar,Piyush, ManoharU. Kalwani,MaqboolDada. 1997.The
impactof waitingtimeguaranteeson customers'waitingexpe-
riences.MarketingSci.16(4)295-314.
Lewis, Michael.2005.A dynamicprogrammingapproachto cus-
tomerrelationshippricing.ManagementSci.51(6)981-994.
Li,Shibo,BaohongSun,AlanMontgomery.2005.Introducingwhat
financialproductto whichcustomerat what time- An empir-
ical analysis of customized and dynamic cross-sellingcam-
paigns. Workingpaper, Carnegie Mellon University,Pitts-
burgh,PA.
Li, Shibo, Baohong Sun, Ronald Wilcox.2006. Cross-sellingse-
quentiallyorderedproducts:Anapplicationtoconsumerbank-
ing services./. MarketingRes.XLII(May)233-239.
Rust,RolandT.,TuckSiongChung.2006.Marketingmodelsof ser-
vice and relationships.MarketingSci.25(6)560-580.
Rust,RolandT., KatherineN. Lemon,ValarieA. Zeithaml.2004.
Returnon marketing:Using customerequityto focusmarket-
ing strategy/. Marketing68(1)109-127.
Sawhney, Mohanbir,Sridhar Balasubramanian,Vish Krishnan.
2004.Creatinggrowth with services.MITSloanManagement
Rev.45(2).
Sullivan, U. Y., Jackie Thomas. 2004. Customer migration:An
empirical investigation across multiple channels. Working
paper,NorthwesternUniversity,Evanston,IL.
Sun,Baohong,ShiboLi.2005.Learningand actingupon customer
information- An empiricalinvestigationof serviceallocations
with off-shorecenters.Workingpaper,CarnegieMellonUni-
versity,Pittsburgh,PA.
Sun, Baohong,YachengSun, Shibo Li. 2005. Pay now and play
later- An empiricalinvestigationof consumerpurchaseand
consumptionbehaviorof online DVD rental.Workingpaper,
CarnegieMellonUniversity,Pittsburgh,PA.
Sun, Baohong,ShiboLi, RongZhou.2006.Adaptivelearningand
"proactive"customerrelationshipmanagement./. Interactive
Marketing.Forthcoming.
Taylor,Gail Ayala,ScottA. Neslin. 2004.The currentand future
sales impactof a retailfrequencyrewardprogram.Working
paper,DartmouthUniversity,Hanover,NH.
Winer,Russell.2001.A frameworkforcustomerrelationshipman-
agement.CaliforniaManagementRev.43(Summer)89-105.
Witten,Ian H., EibeFrank.2005.Data mining:Practicalmachine
learningtools and techniques.Elsevier.
Xie,Jinhong,StevenShugan.2001.Electronictickets,smartcards,
andonlineprepayments:Whenandhow to advancesell.Mar-
ketingSci.20(Summer)219-243.