Glossary
Cohort Analysis
Tracking how user groups acquired together behave over time.
By Amit Tyagi, Fitoor Capital · AletheiaAI Glossary
Definition
Cohort analysis groups users by their acquisition date (week, month, or quarter) and tracks their behavior—retention, spending, churn—as they age. Instead of averaging all users, you see Month 0 retention, Month 3 retention, Month 6 retention for each cohort separately.
This reveals product truth that vanity metrics hide. A SaaS app claiming 80% monthly active users means nothing if Cohort Jan-2024 drops to 15% retention by month 3, while Cohort Jul-2024 holds 60% at month 3. The latter shows genuine product-market fit; the former shows poor retention masked by new user growth.
Cohort analysis unmasks churn seasonality, feature impact, pricing changes, and acquisition quality. If paid cohorts outperform organic cohorts after month 2, your product hooks paid users better. If retention flatlines at month 4 across all cohorts, you have a structural problem.
In spreadsheets, create a matrix: rows = cohorts, columns = age (months since signup), cells = % retained or ARPU. Track weekly for early-stage startups, monthly after Series A. Most investors ask for this before term sheets.
India Context
Indian SaaS founders often chase user growth to impress VCs, but domestic investors increasingly demand cohort retention as the core metric. NASSCOM's 2023 SaaS report found that Indian companies with 50%+ month-3 retention attracted 2.3x more follow-on funding than those below 30%. Blume Ventures, Sequoia India, and Lightspeed India now request cohort analysis as standard due diligence before Series A.
RBI fintech regulations (RBI Master Circular, 2023) don't mandate cohort analysis, but compliance and churn interact: if user lockout happens post-regulation, cohorts acquired before and after the rule shift will show divergent retention patterns. Tracking this separately prevents false optimization.
India's high smartphone adoption (920M users, 2024) but shallow engagement means cohort retention is often 10-20% lower than US benchmarks by month 3. A 40% month-3 retention in India is strong; in the US, it signals danger.
Example
Juno Education (ed-tech startup, Bengaluru) used cohort analysis to fix CAC payback. Their overall retention looked 45%, but cohort breakdown revealed: organic cohorts (free signups) dropped 25% by month 2, while paid trial cohorts held 65%. They shifted marketing budget away from free channels, reduced CAC spend by 30%, and improved unit economics by month 4. Investors saw the pivot via cohort metrics, not surface-level retention lies.
Without cohort analysis, Juno's founder would have seen 45% and blamed product. With it, he saw the acquisition channel was the problem.
Frequently Asked Questions
Related Terms
Apply what you've learned
See this term at work on real Indian companies.
AletheiaAI checks market narratives against the filings behind them — screener, company disclosures, and sector reports across India’s listed companies, free.