What is cohort analysis: Answer Guide 2027
Cohort analysis groups customers by when they joined (or another shared characteristic) and tracks each group's behavior over time — revealing retention, monetization, and payback trends that aggregate numbers hide. In a cohort analysis interview, explain the method, what it reveals, and give the classic retention-curve example.
Cohort Analysis Interview Questions: What They Test
Interviewers want to know why averages lie. Total revenue can grow while each successive cohort performs worse — masked by a growing top of funnel. Cohort tables expose this: rows are cohorts (e.g., January signups), columns are months since joining, and cells show retention or revenue per cohort.
The classic read is the retention curve — what percentage of each cohort is still active after 1, 3, 6, 12 months. Flattening curves signal product-market fit; steep, parallel declines signal a leaky bucket. In lending and credit, cohort (vintage) analysis tracks default curves by origination period, which is how underwriting quality is actually monitored.
How to Answer a Cohort Analysis Interview Question
- Define it. "Group users by shared start date or trait, then track each group's metrics over its own lifetime."
- Explain what it reveals. "Whether newer cohorts are better or worse than older ones — something blended averages conceal."
- Describe the retention curve. "Plot percent retained by months-since-joining; flattening means stickiness, steep drops mean churn problems."
- Name a second use case. "In credit, vintage analysis tracks defaults by origination cohort to judge underwriting quality."
Common Mistakes in Cohort Analysis Interview Answers
- Confusing cohorts with segments. Cohorts share a time-based entry point; segments share attributes — the time dimension is what makes it cohort analysis.
- Reading blended metrics. Quoting overall retention when asked about cohorts misses the entire point of the method.
- Ignoring survivorship. Later periods only include surviving cohorts — always note the shrinking sample.
Cohort thinking is a strong signal in any data-adjacent interview — it shows you distrust averages on principle.
Keep Reading
- investment banking rejected what next
- binary search tree interview question
- Goldman Sachs Risk Video Interview: Evaluate a News Event
- Goldman Sachs Spring Week 2027: Application Guide for First-Years
FAQ
Q: What is a cohort in cohort analysis? A: A group of users or loans sharing a common characteristic — most often the time period when they joined or were originated.
Q: What does a cohort retention curve show? A: The percentage of each cohort still active or paying at successive intervals after joining, revealing stickiness and churn patterns.
Q: How is cohort analysis used in lending? A: Vintage analysis tracks delinquency and default curves by origination period to assess underwriting quality over time.
Q: Why is cohort analysis better than aggregate metrics? A: Aggregates can mask deteriorating unit economics behind growth — cohorts isolate whether each generation of customers is actually healthy.
Preparing for Goldman Sachs's interview? Our 2027 Goldman Sachs Video Interview, Coding & Math Technical Assessment Answers has practice questions and answers — $79 one-time, instant download.
