| Course | DAA-7030 Data Analytics and Visualization with Capstone Project |
|---|---|
| Module | Module 2 |
| Paper type | Written assignment 1: model, solutions and process improvement for a scenario |
| Length | About 510 words, 4 pages |
| Format | APA 7 student paper |
| School | Thomas Edison State University |
| Program | MBA in Data Analytics |
| Updated | October 2026 |
Free sample paper for DAA 7030 Module 2
Who Will Cancel, and Who Can Be Saved? A Data Mining Model and Process Fix for a Fitness Chain
Student Name
MBA in Data Analytics, Thomas Edison State University
DAA-7030: Data Analytics and Visualization with Capstone Project
Mentor Name
Month Day, Year
Who Will Cancel, and Who Can Be Saved? A Data Mining Model and Process Fix for a Fitness Chain
The Scenario
Monthly cancellations at our six studios rose from 3.1 to 4.4 percent of members over the past year. Each lost member is worth about $1,300 in revenue over the following year. The owners want to know who is likely to cancel and what to do about it.
The Model
Fayyad et al. (1996) describe knowledge discovery as a sequence: understanding the business goal, selecting and preparing data, transforming it, mining it and interpreting the results, with loops back as needed. Following those steps, I defined the target as a member canceling within the next 30 days and assembled 18 months of records for 4,200 members: class bookings by week, no-shows, payment failures, tenure, membership type and home studio. Preparation found that 9 percent of members had duplicate records from studio transfers, which were merged.
A logistic regression on the first twelve months, tested on the last six, identified the strongest predictors: a drop of more than half in weekly bookings over four weeks, a failed payment in the past 60 days and tenure under four months. The model ranked members well enough that the top tenth by predicted risk contained 41 percent of actual cancellations.
The Solutions
The obvious solution is to contact the highest-risk members. Ascarza (2018) found, in field experiments, that customers at highest risk of leaving were not necessarily the ones whose behavior a retention offer changed, and that targeting customers by their predicted response to the intervention retained more customers than targeting by risk alone. Some high-risk members have already decided; some moderate-risk members can be persuaded.
I therefore propose a test rather than a campaign. For two months, members in the top three risk deciles will be randomly assigned to receive a personal message from their favorite instructor with a free guest pass, or no contact. Comparing cancellations between the groups, within each risk decile, will show where outreach changes behavior.
| Step | What was done |
|---|---|
| Business goal | Predict 30-day cancellation; find who can be retained |
| Data | 4,200 members, 18 months of bookings, payments and tenure |
| Preparation | Merged duplicates from studio transfers |
| Mining | Logistic regression, tested on the last six months |
| Interpretation | Booking drop, failed payment and short tenure predict cancellation |
| Action | Randomized outreach test by risk decile |
Process Improvement
The model cannot learn why members leave because the chain does not record reasons. Front desk staff process cancellations by email without a reason code. Loshin (2012) stresses that BI depends on data quality at the point of capture. The cancellation form will add a required reason field with seven choices, and staff will offer a brief exit conversation. Within six months these reasons can be added to the model and to the owners' monthly review.
Conclusion
The model predicts who is likely to cancel, the outreach test will show who can be kept and the new reason codes will improve both over time. Together they turn a rising cancellation rate into a managed process.
References
Ascarza, E. (2018). Retention futility: Targeting high-risk customers might be ineffective. Journal of Marketing Research, 55(1), 80-98. https://doi.org/10.1509/jmr.16.0163
Fayyad, U., Piatetsky-Shapiro, G., & Smyth, P. (1996). The KDD process for extracting useful knowledge from volumes of data. Communications of the ACM, 39(11), 27-34. https://doi.org/10.1145/240455.240464
Loshin, D. (2012). Business intelligence: The savvy manager's guide (2nd ed.). Morgan Kaufmann.
What the DAA 7030 Module 2 instructions ask for
The first of DAA-7030's two three-page written assignments asks you to put forth the model, solutions and process improvement for a given scenario; the two together carry 10 percent, and the course text is Loshin's Business Intelligence. Module 2, Steps to Data Mining, supplies the method. Expect to define the business problem precisely, walk through the data mining steps you used or would use, report what the model found, propose solutions that follow from it and improve the business process that produces or uses the data. Keep to three pages in APA format, so every section must be concise and specific. Report one figure that shows how well the model works. Propose a solution you can test, not only a model.
How this DAA 7030 Module 2 example is built
The paper opens with the scenario in figures: cancellations rising from 3.1 to 4.4 percent and $1,300 lost per member. Fayyad and colleagues' knowledge discovery steps structure the model, from defining the target through data preparation to a logistic regression tested on later months. Ascarza's field experiments change the solution from targeting the riskiest members to a randomized outreach test by risk decile. A six-row table summarizes each step. The process improvement adds required cancellation reasons, citing Loshin on data quality at capture. A margin note explains why testing on later months matters. A six-row table walks through each step from goal to action.
Where the marks sit in the DAA 7030 Module 2 rubric
Mentors grade DAA-7030 written assignments on correct use of data mining concepts, a clear model with reported results, solutions that follow from the analysis, a sensible process improvement, research support and APA writing within the length. Testing on later months and giving one accuracy figure earns credit. Using research to refine the solution, rather than acting on predictions naively, shows sophistication. Fixing the process that generates data shows understanding of BI as a system. Concise writing matters because the assignment is limited to three pages. Mentors also credit a process fix that improves the data future models will use.
DAA 7030 Module 2 help: mistakes that cost marks
Students lose marks by describing the data mining steps without applying them, by reporting a model without testing it or by proposing to target everyone the model flags. Others forget the process improvement the assignment requires. Set the outcome to predict, take each step with real or realistic records, test on months the model never saw, design a measurable fix and repair how the data are gathered. A free first draft of your DAA-7030 Module 2 written assignment can be built on your scenario. Keep the model's performance figure in plain words, such as the share of cancellations caught in the top tenth. Fix how the data are captured, since better data improve every later model.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official Thomas Edison State University document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.
More DAA 7030 and MBA in Data Analytics sample papers
- DAA 7030 Module 1: Discussion: What Business Intelligence Means for a Fitness Studio Chain
- DAA 7030 Module 3: Midterm Paper: Collecting, Mining and Presenting Public Data on Where to Open the Next Studio
- DAA 7030 Module 4: Written Assignment: Choosing Analytics Software for a Small Chain's First BI Project
- DAA 7030 Module 5: Discussion: One BI Platform for Classes, Staff and Money at a Fitness Chain
- DAA 7030 Module 6: Capstone Project: A Data-Driven Marketing Plan for a Seventh Fitness Studio
- MST 7000 Module 5: Written Assignment: Are Carrier B's Deliveries Really Faster?
- BFO 7010 Module 7: Written Assignment: Identifying and Fitting a Seasonal ARIMA Model
- DAM 7020 Module 1: Discussion: Descriptive, Predictive or Prescriptive Analytics for a Tire and Auto Service Chain?
DAA 7030 Module 2 questions, answered
Where can I find a free DAA 7030 Module 2 sample paper?
The written assignment above builds a cancellation model and process fix for a fitness studio chain.
What are the steps of knowledge discovery in data mining?
Understanding the goal, selecting and preparing data, transforming it, mining it and interpreting the results, with loops back.
Should retention efforts target the highest-risk customers?
Ascarza found the highest-risk customers are not always the most responsive, so targeting by predicted response can retain more.
Why test a model on later data?
Testing on data the model was not built on shows how it would perform on future cases.
How long are the DAA-7030 written assignments?
Each is three pages in APA format, and the two together are worth 10 percent.