Peeking Inside the Wallet
Skip to content
Marketing Finance & Accounting Nov 1, 2012

Peek­ing Inside the Wallet

How much do con­sumers spend on their cred­it cards?

Based on the research of

Yuxin Chen

Joel H. Steckel

Recent advances in tech­nol­o­gy and mar­ket­ing sci­ence have enabled com­pa­nies to employ pow­er­ful new strate­gies to tar­get their mar­ket­ing efforts and max­i­mize cus­tomer value.

Add Insight
to your inbox.

We’ll send you one email a week with content you actually want to read, curated by the Insight team.

But they still face a frus­trat­ing obsta­cle: they often have incom­plete infor­ma­tion about their cus­tomers’ behav­ior. They know what trans­ac­tions their cus­tomers have engaged in with their com­pa­ny, but do not know about their cus­tomers’ trans­ac­tions with competitors.

For exam­ple, a cred­it card com­pa­ny knows what each of its cus­tomers buys with its card, but it does not know about pur­chas­es made with oth­er cards, cash, or checks. Yux­in Chen, pro­fes­sor of mar­ket­ing at the Kel­logg School of Man­age­ment, and Joel H. Steck­el, pro­fes­sor at New York Uni­ver­si­ty, have cre­at­ed a mod­el that allows com­pa­nies to com­bine infor­ma­tion from oth­er sources with the datasets they already have and gain use­ful insights into cus­tomer behavior.

For exam­ple, a cred­it card com­pa­ny might be con­sid­er­ing a mar­ket­ing cam­paign to entice cus­tomers to use its card more fre­quent­ly to buy gro­ceries. To effec­tive­ly tar­get such a cam­paign, it needs to know how often each cus­tomer uses the card to buy gro­ceries and how much they spend, then deter­mine the company’s share of wal­let” — the per­cent­age of the customer’s total gro­cery spend­ing that is done with that spe­cif­ic cred­it card.

If the card has a low share of wal­let and the tar­get­ed con­sumer goes to the gro­cery store very fre­quent­ly, the pro­mo­tion has a good chance of suc­ceed­ing. On the oth­er hand, if the card has a high share of wal­let and the tar­get­ed con­sumer eats out most of the time and rarely goes gro­cery shop­ping, the pro­mo­tion is like­ly to fail,” Chen and Steck­el explain in a recent paper.

If the card has a low share of wal­let and the tar­get­ed con­sumer goes to the gro­cery store very fre­quent­ly, the pro­mo­tion has a good chance of suc­ceed­ing.” — Chen and Steckel

Because com­pa­nies do not see the whole pic­ture, it is dif­fi­cult for them to pre­cise­ly tar­get incen­tives and pro­mo­tions to the cus­tomers who have the high­est poten­tial val­ue. In their paper, Chen and Steck­el report on how their new mod­el helps com­pa­nies fill in these infor­ma­tion gaps.

The Incom­plete Infor­ma­tion” Puz­zle
The researchers obtained data on 350 con­sumers who col­lec­tive­ly made 8,348 gro­cery pur­chas­es using a cer­tain Visa card. They com­bined this data with infor­ma­tion pre­vi­ous­ly estab­lished by oth­er research: the sto­chas­tic, or ran­dom, pat­tern of how fre­quent­ly peo­ple shop for groceries. 

We know that the pat­tern of inter­pur­chase times has to fol­low cer­tain dis­tri­b­u­tions on an aggre­gate lev­el — that gro­cery pur­chas­es fol­low a cer­tain dis­tri­b­u­tion,” Chen explains. We don’t real­ly need to know the para­me­ters of the dis­tri­b­u­tion. The pat­tern is what we need to know, not the value.”

Inter­pur­chase times form the core of the incom­plete infor­ma­tion prob­lem; i.e., miss­ing pur­chase occa­sions in the data can ren­der the observed inter­pur­chase times very dif­fer­ent from the under­ly­ing truth,” Chen and Steck­el write in their paper.

From the com­bi­na­tion of card­hold­er pur­chase data and knowl­edge of the pat­tern of inter­pur­chase times, Chen and Steck­el were able to cre­ate a math­e­mat­i­cal mod­el that deduced how con­sumers buy gro­ceries using oth­er means of pur­chase — oth­er cred­it cards, cash, and checks. The share-of-wal­let infor­ma­tion they esti­mat­ed on the 350 card­hold­ers sug­gest­ed that, over­all, these cus­tomers used the cards involved in the study for about one out of five gro­cery pur­chas­es, but there was a wide range in the usage pat­terns (Fig­ure 1). For exam­ple, one cus­tomer used the card for more than 90 per­cent of his pur­chas­es. This dif­fer­ence in poten­tial cus­tomer val­ue has spe­cif­ic impli­ca­tions for help­ing com­pa­nies tar­get their promotions.

Chen2012_-Fig1.gif

Fig­ure 1. Per­cent­ages of gro­cery pur­chas­es that 350 cred­it card cus­tomers made with the study’s fic­ti­tious Princess Bank” VISA card.

One per­son may spend $5,000 with you, but that per­son is already spend­ing 90 per­cent of his mon­ey with you. Anoth­er per­son may spend $5,000 with you, but may spend $10,000 alto­geth­er,” Chen notes. With­out this method, you would see these two peo­ple in the same way, but now we see that the sec­ond per­son has a lot of upside poten­tial for you. You may want to offer a pro­mo­tion or incen­tive to the sec­ond per­son, not to the first person.”

Tra­di­tion­al­ly, he says, a com­pa­ny would sim­ply look at the sum spent with it to decide who was valu­able and who was not. The new mod­el allows com­pa­nies to dis­en­tan­gle cat­e­go­ry spend­ing and share of wal­let to pro­vide a bet­ter pic­ture of cus­tomer behavior.

Accord­ing to Chen, many writ­ers have cit­ed the 8020 rule,” mean­ing that 20 per­cent of cus­tomers con­tribute around 80 per­cent of most com­pa­nies’ total rev­enue. How­ev­er, his research revealed that the rule is clos­er to 80 – 59 at the cat­e­go­ry lev­el for gro­ceries, mean­ing that 59 per­cent of cus­tomers con­tribute 80 per­cent of rev­enue in the gro­cery category.

Broad Applic­a­bil­i­ty
Chen has already applied a sim­i­lar idea on behalf of oth­er com­pa­nies. For exam­ple, in con­sult­ing with a Chi­nese air­line, he iden­ti­fied peo­ple who trav­eled by air fre­quent­ly but who rarely flew on that spe­cif­ic air­line. His find­ings enabled the air­line to design incen­tives to entice those trav­el­ers to choose their air­line more frequently.

The mod­el might yield inter­est­ing results for oth­er retail­ers, Chen and Steck­el believe. What frac­tion of books does a typ­i­cal Ama​zon​.com cus­tomer buy through Ama­zon?” they ask in their paper. What is Home Depot’s [share of wal­let] with respect to its customers?”

They antic­i­pate that com­pa­nies will be able to apply the mod­el if they have a com­plete set of cus­tomer pur­chase data, if the prod­uct cat­e­go­ry is one in which pur­chase behav­ior is sta­tion­ary over a rel­a­tive­ly long peri­od (for exam­ple, pur­chas­es of gaso­line and telecom­mu­ni­ca­tions ser­vices are more pre­dictable than pur­chas­es of homes and auto­mo­biles), and if the com­pa­ny knows the pat­tern of inter­pur­chase times. The researchers’ next step will be to see if the mod­el can deliv­er on its promise to help com­pa­nies reli­ably increase the prof­itabil­i­ty of their cus­tomer bases.


Relat­ed read­ing on Kel­logg Insight

When Mem­o­ry Serves: Cus­tomers Bet­ter Remem­ber Low Prices

Mem­ber­ship Has Its Pun­ish­ments: Loy­al­ty Pro­grams Dis­suade Firms from Pric­ing Prod­ucts Competitively

Featured Faculty

Yuxin Chen

Member of the Department of Marketing faculty from 2009 to 2013

About the Writer

Beverly A. Caley, JD, is an independent writer based in Corvallis, Oregon, who concentrates on business, legal, and science topics.

About the Research

Chen, Yuxin, and Joel H. Steckel. 2012. “Modeling Credit Card Share of Wallet: Solving the Incomplete Information Problem.” Journal of Marketing Research, 49(5): 655-669.

Read the original

Suggested For You

Most Popular

Organizations

How Are Black – White Bira­cial Peo­ple Per­ceived in Terms of Race?

Under­stand­ing the answer — and why black and white Amer­i­cans’ respons­es may dif­fer — is increas­ing­ly impor­tant in a mul­tira­cial society.

Leadership

Why Warmth Is the Under­ap­pre­ci­at­ed Skill Lead­ers Need

The case for demon­strat­ing more than just competence.

Most Popular Podcasts

Careers

Pod­cast: Our Most Pop­u­lar Advice on Improv­ing Rela­tion­ships with Colleagues

Cowork­ers can make us crazy. Here’s how to han­dle tough situations.

Social Impact

Pod­cast: How You and Your Com­pa­ny Can Lend Exper­tise to a Non­prof­it in Need

Plus: Four ques­tions to con­sid­er before becom­ing a social-impact entrepreneur.

Careers

Pod­cast: Attract Rock­star Employ­ees — or Devel­op Your Own

Find­ing and nur­tur­ing high per­form­ers isn’t easy, but it pays off.

Marketing

Pod­cast: How Music Can Change Our Mood

A Broad­way song­writer and a mar­ket­ing pro­fes­sor dis­cuss the con­nec­tion between our favorite tunes and how they make us feel.