SaaS Churn &
Retention Analysis
A SaaS CFO wants to know why customers leave and what to do about it. You'll run TWO independent investigations using different methods — cohort retention and behavioural predictors — then make the meta-call: which one is actually more actionable. Most analysts run every analysis they can think of. Mature analysts run several and recommend one. This project teaches the second skill explicitly.
What you'll learn
- —Compute MRR and churn rate from raw subscription data — defensibly
- —Run a cohort retention analysis and articulate what its shape reveals
- —Identify behavioural churn predictors using event-log comparison
- —Make the meta-call: which analytical lens is more actionable, and why
- —Communicate that meta-call to a CFO in plain language with numbers
The 5 milestones
Each milestone is reviewed before you advance.
- 01
Set up your project repo
Before any analytical work: create a public GitHub repository, push the standard skeleton, and paste the URL into your workspace.
- 02
Foundation — definitions, MRR, topline churn
Map the SaaS data, lock down 'active' and 'churned' definitions, and compute the headline numbers.
- 03
Investigation A — cohort retention
First independent investigation: build cohort retention curves and articulate what they reveal.
- 04
Investigation B — behavioural predictors
Second independent investigation: identify what behavioural signals differ between churned and retained customers.
- 05
Synthesize — which lens matters more, and what to do
The meta-skill milestone: pick the more actionable investigation, argue for it, and ship the portfolio piece.
Reading & references
Build your portfolio & get certified
Want more projects like this? Join our Data Analyst Work Experience Program to complete 5+ guided industry projects, gain real experience, and earn an internship certificate.