QNT 375 Week 1 Framing a Business Problem Example

Why Are Members Canceling? Framing a Data Analytics Problem for a Colorado Car Wash Chain

[Student Name]

University of Phoenix

QNT/375: Business Data Analytics

Week 1 Assignment

[Instructor Name]

[Date]

Summit Shine Car Wash, its locations, members and figures are composites written for a model paper.

Summit Shine Car Wash, a composite company, runs 22 express car washes across the Denver metropolitan area. Most revenue comes from unlimited wash memberships: about 61,000 members pay $25 to $45 a month, depending on the plan, and wash as often as they like. Memberships made the business predictable and grew quickly for five years. Over the past three quarters, however, the monthly cancellation rate rose from 6.1 to 8.4 percent, and membership growth stalled even as new sign-ups held steady. The chief executive has asked the finance and operations team to "figure out what is going on with cancellations." This paper frames that request as an analytics problem.

Symptoms and Explanations

Managers offer several explanations. The marketing director blames a competitor that opened five locations with lower prices. Site managers blame long lines on weekends, which have grown with membership. The chief financial officer notes that prices rose by $3 a month last spring. Others point to a dry winter with fewer snowstorms, which reduced the need to wash salt off cars. Each explanation may contain truth, but none has been tested, and each implies a different response.

The rising cancellation rate is a symptom. The explanations are hypotheses. The underlying problem must be stated in terms of a decision.

The Decision

Davenport and Harris (2017) argue that analytics creates value when it informs specific decisions rather than producing reports for their own sake. Summit Shine's leaders face a concrete decision: the board has approved a $600,000 retention budget for next year. They must decide how to spend it: price discounts, capacity at busy sites, targeted offers to at-risk members, service improvements or marketing. Spending it evenly across all members would waste much of it on members who would stay anyway.

Problem Statement

Summit Shine needs to identify which members are most likely to cancel and the main reasons they cancel, so that it can direct its retention budget to the members and actions with the greatest effect on cancellation.

Research Questions

RQ1: How do cancellation rates differ by location, plan type, member tenure and price paid?

RQ2: Is cancellation associated with members' recent washing frequency, wait times at their home location or proximity to a competitor?

RQ3: Did cancellation rates change after the price increase and after the competitor's openings, controlling for season?

RQ4: Can member data predict which current members will cancel in the next 60 days with useful accuracy?

Hypotheses

H1: Members whose washing frequency declined in the past two months are more likely to cancel.

H2: Members whose home location is within two miles of a competitor's site cancel at higher rates.

H3: Members at locations with average weekend wait times above 15 minutes cancel at higher rates.

H4: Cancellation rates rose more among members on the lowest-priced plan after the price increase.

A member who stops washing in March is telling the company something in April's data before she cancels in May.

What a Retained Member Is Worth

The stakes justify careful framing. The average member pays about $32 a month and, before the recent rise in cancellations, stayed about 22 months, for a lifetime value of roughly $700. At 61,000 members, each percentage point of monthly cancellation represents about 610 members a month, or about $19,500 in monthly revenue that compounds as lost members stop paying. Reducing the cancellation rate from 8.4 back toward 6 percent would preserve roughly $3.5 million in revenue in the first year alone, and more as retained members keep paying. Against that, the $600,000 retention budget is modest, but only if it is spent on the right members.

What the Analysis Will Not Do

Framing also sets limits. The study will not redesign pricing or plan structures, decide whether to open new locations or evaluate the competitor's business. Those are important questions, but they belong to other decisions. Keeping the scope tied to the retention budget prevents the project from growing into a general review of the business that takes months and answers nothing clearly.

Explain or Predict?

Shmueli and Koppius (2011) distinguished explanatory analysis, which tests causal hypotheses, from predictive analysis, which aims to forecast outcomes for new observations, and argued that the two require different modeling choices and are judged by different criteria. RQ1 through RQ3 are explanatory: they ask why members cancel, which guides the type of retention action. RQ4 is predictive: it asks who will cancel, which guides where to spend. Summit Shine needs both.

Data Needed

Provost and Fawcett (2013) argue that firms gain from letting analysis, rather than hunches, guide choices, but only when the right data are gathered and brought together. Summit Shine has more data than it uses:

Membership system: plan, price, start date, cancellation date and stated reason when given.

Wash logs: every wash by member, location and time, from license plate readers.

Site operations: wait times estimated from queue cameras and wash counts.

External: competitor locations and opening dates, weather records.

New collection: a short survey of recently canceled members and a sample of current ones.

A Plan for the Weeks Ahead

The project will follow the course's sequence. Week 2 will design the data collection, deciding which questions existing records can answer and where a survey is needed. Week 3 will test the data's validity and reliability before any conclusions are drawn. Week 4 will analyze patterns, test the hypotheses and build a prediction model. Week 5 will present conclusions and a recommended budget to the board. The team has six weeks before the board meeting, which is enough time if the data work begins immediately.

Who Will Use the Results

The chief executive and board will use the findings to allocate the retention budget; site managers will use location results to plan staffing; marketing will use predictions to target offers.

Conclusion

"Figure out what is going on with cancellations" becomes an answerable problem once it is tied to a decision, how to spend $600,000 on retention, and broken into research questions and hypotheses. Distinguishing explanatory and predictive goals and identifying data sources sets up the research design in Week 2.

References

Davenport, T. H., & Harris, J. G. (2017). Competing on analytics: The new science of winning (Updated ed.). Harvard Business Review Press.

Provost, F., & Fawcett, T. (2013). Data science and its relationship to big data and data-driven decision making. Big Data, 1(1), 51-59. https://doi.org/10.1089/big.2013.1508

Shmueli, G., & Koppius, O. R. (2011). Predictive analytics in information systems research. MIS Quarterly, 35(3), 553-572. https://doi.org/10.2307/23042796