BUS 499 Module 6 Analysis and Findings Example

A Stranger at the Door and a Bill With No Bugs: Findings on Why Customers Cancel Quarterly Pest Control Plans

Student Name

Business Administration Program, Aspen University

BUS 499: Senior Capstone

Instructor Name

Month Day, Year

A Stranger at the Door and a Bill With No Bugs: Findings on Why Customers Cancel Quarterly Pest Control Plans

This module reports what the capstone study found about why Prairie Shield's homeowners leave. The prepared data include records for 3,186 households, of whom 1,054 ended their plan inside a year, and interviews with 19 former customers. The findings are presented in four parts: cancellation rates by group, a regression testing the four hypotheses, interview themes and a comparison of the two kinds of evidence.

Cancellation Rates by Group

The table shows the share of each group that canceled within twelve months.

The differences are large. Customers billed per visit canceled at almost twice the rate of customers on automatic payment, and two in three customers with an unresolved complaint canceled. Customers whose complaint was resolved quickly canceled at about the same rate as customers with no complaint.

Testing the Hypotheses

Because customers fall into several groups at once, the rates above could overlap; for example, first-year customers might be more likely to pay per visit. Logistic regression estimates each factor's association with cancellation while holding the others constant (Hosmer et al., 2013). All predictors were entered together.

Hypothesis 1, that per-visit billing is associated with cancellation, was supported: the odds of canceling were about twice as high for per-visit customers after accounting for the other factors. Hypothesis 2, about technician continuity, was supported, with odds about 2.3 times higher when technicians changed. Hypothesis 3 was supported for unresolved complaints, which had the strongest association of any factor, but not for complaints resolved within 48 hours, whose confidence interval includes 1. This matches the research of Smith et al. (1999), which found that a recovery matched to the failure restores much of the satisfaction lost. Hypothesis 4, about first-year customers, was supported. The price increase, which the owners had considered the most likely cause, showed no clear association once other factors were included.

How the Factors Combine

Because the factors operate together, it helps to look at customers who carry several at once. Among customers billed per visit who also saw changing technicians, 58% canceled within twelve months. Among customers on automatic payment who kept the same technician, only 15% did. First-year customers with an unresolved complaint were the group most likely to leave, with nearly three in four canceling. These combinations matter for the recommendations, because a change that affects one factor may help most among customers who carry others as well.

Price and Season

The two control variables behaved differently from the owners' expectations. Customers who enrolled at $129 canceled at 35%, compared with 31% at $119, but the difference disappeared in the regression, largely because the higher price arrived at the same time as the new routing software and a wave of new technicians. Season of enrollment showed no clear pattern; customers who joined during spring insect season were no more likely to stay than those who joined in winter.

Sensitivity Check

Because complaint records were incomplete for January through April 2023, the regression was repeated without customers who started in those months. The odds ratios changed by no more than 0.2, and every conclusion above held, so the gap in early records does not appear to have distorted the results.

What Former Customers Said

Thematic analysis of the 19 interviews, following the phases described by Braun and Clarke (2006), produced four themes.

A stranger every time, raised by 11 participants. Customers described not knowing who would arrive. One said, "I used to have a guy who knew where the dog was. Then it was somebody new every quarter."

The bill shows up when nothing is wrong, raised by 9. Customers on per-visit billing described the quarterly invoice as a moment of reconsideration: "It came in January, there hadn't been a spider in months, and I thought, why am I paying this?"

Nobody came back, raised by 7. Customers who called about pests between visits described waiting days or getting no return visit at all.

I thought one treatment would do it, raised by 5. Some first-year customers did not understand that the plan was preventive and expected a single treatment to solve the problem permanently.

Only 3 participants named price as their main reason for leaving.

Comparing Records and Voices

The two kinds of evidence point in the same direction on every factor, and the interviews explain why the patterns in the records appear.

What the Findings Cannot Show

The records show associations, not causes. Customers who chose automatic payment may differ from others in ways the company does not record, such as stability of income or satisfaction at the outset. The interview sample was chosen for variety and is too small to estimate how common each reason is among all former customers. These limits do not undermine the findings, but they shape how confidently recommendations can be made.

Conclusion

Cancellation at Prairie Shield is most strongly associated with unresolved complaints, changing technicians, per-visit billing and being in the first year, and not clearly associated with the price increase. Former customers' accounts match and explain each pattern. The next module turns these findings into recommendations.

References

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. https://doi.org/10.1191/1478088706qp063oa

Hosmer, D. W., Jr., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (3rd ed.). Wiley.

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