Glossary /
Cohort analysis
Written by Peko Research Team.Last updated: 09/15/2026.
Cohort analysis groups customers by a shared starting event (usually their first visit month) and tracks how each group's behaviour evolves over time.
Published: 05/01/2026
Updated September 2026 — Cohort analysis groups customers by a shared starting event (usually their first visit month) and tracks how each group's behaviour evolves over time. In F&B, the most useful cohort is the first-visit month. You plot what % of each month's first-time guests are still visiting at month 1, 2, 3, etc.
Quick facts
- Definition
- Cohort analysis groups customers by a shared starting event (usually their first visit month) and tracks how each group's behaviour evolves over time.
- Why it matters
- In F&B, the most useful cohort is the first-visit month. You plot what % of each month's first-time guests are still visiting at month 1, 2, 3, etc. The shape of the curve tells you exactly when most churn happens — usually a steep drop between month 0 and month 1.
- Worked example
- January cohort: 100% in month 0, 35% in month 1, 22% in month 3, 15% in month 6. After launching automated win-back, the April cohort tracks at 50% / 38% / 28% — a clear lift attributable to the new flow.
- Related terms
- Customer retention rate, Customer churn, RFM analysis
In F&B, the most useful cohort is the first-visit month. You plot what % of each month's first-time guests are still visiting at month 1, 2, 3, etc. The shape of the curve tells you exactly when most churn happens — usually a steep drop between month 0 and month 1.
Comparing cohorts before and after a change (new menu, loyalty launch, staff training) is the cleanest way to measure retention impact, because each cohort acts as its own control group.
Worked example
January cohort: 100% in month 0, 35% in month 1, 22% in month 3, 15% in month 6. After launching automated win-back, the April cohort tracks at 50% / 38% / 28% — a clear lift attributable to the new flow.
Numbers on this page: named sources are listed below; figures without a named source are PEKO merchant data, Vietnam, Jan 2024 – Jun 2026, or PEKO estimates where modelled. See the datasets behind these numbers.
FAQ
What tools do I need for restaurant cohort analysis?
Any CRM with POS sync can compute it — PEKO ships cohort retention curves out of the box. Spreadsheets work too if you can export customer + visit data.
How many cohorts should I compare to trust the result?
Minimum three consecutive monthly cohorts before a change and three after — fewer than that and seasonality or a single bad week can swing the curve. Six-on-six is the gold standard for attribution decisions.
What does a healthy F&B cohort retention curve look like?
A steep drop from month 0 to month 1 (typical: 100% → 30–45%), then a flattening tail that holds 15–25% by month 6. A curve that keeps falling steeply past month 2 signals the post-first-visit experience is the leak, not acquisition.
Sources
The definitions and figures on this page reference the sources below:
Related terms
People also read
Answer
How do I reduce customer churn in my restaurant?
Capture every guest's contact at first visit, segment by recency and frequency, then trigger an automated win-back the moment a regular's silence breaks their normal cadence — typically lifts retention 8–15 percentage points in 90 days.
Term
Loyalty death spiral
The loyalty death spiral is a self-reinforcing 6-step loop that quietly kills a points program from the inside. It starts with cashiers skipping enrolment on busy shifts, drops new-member growth to near zero, hides point balances from members so they forget the program exists, and ends with the owner concluding 'loyalty doesn't work here' and shutting it off.
Term
Win-back campaign
A win-back campaign is a targeted message — usually with an incentive — sent to customers who have stopped visiting, with the goal of triggering a return visit.