BUSI 4004 Week 3 Lost Customer Retention Analysis Example

Why a Two-Year Weekly Regular Stopped Coming: A Retention Analysis for a Neighborhood Dry Cleaner That Sets One Lost Customer's Ticket History Against a Simple Cohort View and Sorts the Evidence Into Proven, Probable and Never Asked

Student Name

Walden University

BUSI 4004: Customer Relationship Management for Small Business

Instructor Name

Month Day, Year

Why a Two-Year Weekly Regular Stopped Coming: A Retention Analysis for a Neighborhood Dry Cleaner That Sets One Lost Customer's Ticket History Against a Simple Cohort View and Sorts the Evidence Into Proven, Probable and Never Asked

The Customer Who Stopped

For two years, a customer at Grand River Cleaners, a composite family dry cleaner in Lansing, Michigan, brought in five or six dress shirts nearly every Monday. His ticket history shows 94 orders between March 2025 and March 2027, an average of $38 an order, or about $3,600 a year. His last order was picked up on March 11, 2027. He has not been back since, and nobody at the shop knows why. The owner noticed only in June, when a clerk mentioned that "the Monday shirts man" had not been in.

This analysis asks why he left and whether the shop could have known sooner. It uses two kinds of evidence: his own ticket history and a simple comparison with other regulars who started at about the same time.

What His Ticket History Shows

Table 1

The Lost Customer's Last Ten Weeks of Orders

Shop and customer are both composite.

Three service failures appear in five weeks: a scorched collar and two pickups at which he waited after being told his order was ready. After the third failure, his orders shrank from five or six shirts to four, then three, then stopped.

The Cohort View

One customer could have left for any reason. To see whether his path fits a pattern, I pulled every customer who first came in during 2025 and made at least 20 visits in their first year, 112 people. I then compared those whose records show a service failure in a given three-month period with those whose records do not.

Table 2

Regulars First Seen in 2025 and Still Active Two Years Later

Active means at least one order in the last 60 days. Failures are those noted in the ticket system: damage, late orders, misfiled pickups.

The lost customer belongs to the last group. Among regulars with two or more failures in a short period, fewer than half were still customers two years on. One failure, handled with a credit, did not usually end a relationship; a cluster of them often did. Smith et al. (1999) found that recoveries satisfy customers best when they fit the type and seriousness of what went wrong, and repeated failures leave less room for any recovery to work.

Sorting the Evidence

Proven by the records: he experienced three failures in five weeks; his orders shrank after the third; he stopped completely four weeks later. Regulars with clustered failures leave at a much higher rate than others.

Probable but not proven: the failures caused his departure. The timing fits, the shrinking orders fit and the cohort pattern fits. But the records cannot exclude other causes, such as a move, a new job or a switch to a work-from-home schedule with fewer dress shirts.

Never asked: nobody contacted him. The shop did not call after the misfiled pickup, did not notice when his orders shrank and did not reach out when he stopped. The single most useful piece of evidence, his own reason, was never collected.

What the Shop Could Have Known Sooner

The signal was in the data weeks before he left. A weekly customer whose order size falls by half over a month, right after a cluster of failures, is a customer at risk. Fader et al. (2005) showed how recency and frequency of purchase, data almost every small business already has, can describe how valuable a customer is likely to be in future. Long tenure alone does not guarantee profit (Reinartz & Kumar, 2000), but this customer was both long-standing and steady, the kind worth a phone call.

Recommendations

The analysis points to three practices. First, a weekly report should flag any regular who has missed three expected visits or whose orders have shrunk sharply, and the owner should call each one. Second, two failures for the same customer within three months should trigger a personal call from the owner, not just a credit at the counter. Third, the shop should simply ask: a short call to this customer now, four months late, might still bring him back and would at least supply the missing evidence.

Limits

The cohort is small, 112 customers, and the failure notes depend on clerks recording them, which they may not always do. The analysis shows a strong association between clustered failures and lost regulars, not a proven cause.

References

Fader, P. S., Hardie, B. G. S., & Lee, K. L. (2005). RFM and CLV: Using iso-value curves for customer base analysis. Journal of Marketing Research, 42(4), 415-430. https://doi.org/10.1509/jmkr.2005.42.4.415

Reinartz, W. J., & Kumar, V. (2000). On the profitability of long-life customers in a noncontractual setting: An empirical investigation and implications for marketing. Journal of Marketing, 64(4), 17-35. https://doi.org/10.1509/jmkg.64.4.17.18077

Smith, A. K., Bolton, R. N., & Wagner, J. (1999). A model of customer satisfaction with service encounters involving failure and recovery. Journal of Marketing Research, 36(3), 356-372. https://doi.org/10.1177/002224379903600305