Skip to main content
Back to blog
cohort analysisrevenue retentionSaaS metricsNRRcustomer analytics

Cohort-Based Revenue Analysis: A Beginner's Guide

Top-quartile SaaS companies retain 95%+ of cohort revenue at month 12. Learn cohort analysis step-by-step with real data tables and retention benchmarks.

T
Team culta
·9 min read

Top-quartile SaaS companies retain 95% or more of each cohort's revenue at the 12-month mark. Bottom-quartile companies retain less than 70%. The difference between those two numbers is the difference between a business that compounds growth and one that constantly refills a leaking bucket. Cohort analysis is how you tell which one you are building.

Aggregate metrics like total MRR growth can hide serious problems. You might be growing 10% month-over-month in total revenue while every cohort of customers is shrinking -- if you are acquiring new customers fast enough, the top-line number looks fine. Cohort analysis strips away that illusion and shows you the raw truth about whether customers stick around and spend more over time.

What Cohort Revenue Analysis Actually Is

A cohort is a group of customers who share a common characteristic -- usually the month they signed up. Cohort revenue analysis tracks how much revenue each group generates over time, compared to what they generated in their first month.

Example: Your January 2026 cohort is every customer who signed up in January. You track how much revenue this group generates in February, March, April, and so on. If they generated $10,000 in January and $9,500 in February, their month-2 retention is 95%.

This is different from looking at total revenue, which mixes together customers from every cohort and makes it impossible to see retention patterns.

Building Your First Cohort Table

Step 1: Group Customers by Signup Month

CohortCustomersStarting MRR
October 202545$8,100
November 202552$9,360
December 202538$6,840
January 202661$10,980
February 202658$10,440
March 202667$12,060

Step 2: Track Revenue by Month Since Signup

This is the core cohort table. Each row is a cohort, each column is a month since signup.

Revenue Retention Table (% of starting MRR):

CohortMonth 1Month 2Month 3Month 4Month 5Month 6
Oct 2025100%92%88%85%84%83%
Nov 2025100%94%91%88%87%--
Dec 2025100%93%90%87%----
Jan 2026100%95%92%------
Feb 2026100%96%--------
Mar 2026100%----------

Step 3: Read the Patterns

This table tells you several things:

  1. Retention is improving: Newer cohorts retain better at each month mark (Oct month-2 was 92%, Feb month-2 is 96%)
  2. Most churn happens early: The biggest drop is month 1 to month 2 (4-8%). After month 3, retention stabilizes around 1-2% decline per month
  3. No cohort exceeds 100%: This means you have net negative retention -- more revenue leaves than expands. For a healthy SaaS business, you want some cohorts to exceed 100% through upsells

Use a net revenue retention calculator to calculate your NRR across all cohorts and compare to benchmarks.

Revenue Retention vs. Logo Retention

These are different metrics and tell different stories:

MetricWhat It MeasuresGood Benchmark
Logo retention% of customers still active95%+ monthly
Gross revenue retention% of starting revenue retained (excludes expansion)90%+ annually
Net revenue retention% of starting revenue retained (includes expansion)100%+ annually

Critical insight: A company can have 95% logo retention but only 80% revenue retention if the customers who churn are disproportionately high-value. Conversely, a company with 90% logo retention can have 120% net revenue retention if remaining customers expand significantly.

Always track both. Logo retention tells you about product satisfaction. Revenue retention tells you about business sustainability.

Worked Example: Logo vs. Revenue Divergence

January 2026 cohort (61 customers, $10,980 MRR):

MonthCustomers RemainingRevenueLogo RetentionRevenue Retention
161$10,980100%100%
258$10,44095.1%95.1%
356$10,64091.8%96.9%
454$10,92088.5%99.5%
553$11,20086.9%102.0%

Logo retention fell to 87%, but revenue retention hit 102% by month 5. The remaining customers are expanding their usage and spending. This is the sign of strong product-market fit -- you are losing some customers, but the ones who stay are getting more value over time.

For a deep dive into calculating and improving these metrics, see cohort analysis for SaaS companies.

Benchmarks by Stage and Segment

Annual Net Revenue Retention Benchmarks

Company StageBottom QuartileMedianTop Quartile
Seed (under $1M ARR)70-80%85-95%100-110%
Series A ($1-5M ARR)80-90%95-105%110-130%
Series B ($5-20M ARR)90-100%105-115%120-140%
Growth ($20M+ ARR)95-105%110-120%130-150%

By Customer Segment

SegmentTypical NRRWhy
Enterprise (over $50K ACV)110-140%Strong expansion, low churn
Mid-market ($10-50K ACV)100-120%Moderate expansion, moderate churn
SMB ($1-10K ACV)80-100%Limited expansion, higher churn
Self-serve (under $1K ACV)70-90%High churn, minimal expansion

If your NRR is below 90%, customer lifetime value is severely impacted. Use a customer LTV calculator to see exactly how retention rates compound into LTV differences.

Advanced Cohort Analysis Techniques

Technique 1: Segment Cohorts Beyond Signup Month

Signup month is the default segmentation, but you can learn much more by creating cohorts based on:

  • Acquisition channel: Do customers from organic search retain better than paid ads?
  • Plan tier: Do enterprise customers retain better than SMB?
  • First action: Do customers who complete onboarding in day 1 retain better?
  • Geography: Do US customers retain differently than international?

Technique 2: Revenue Waterfall by Cohort

Break each cohort's revenue change into components:

ComponentDefinitionWhat It Tells You
Retained revenueRevenue from customers at same priceBase retention health
Expansion revenueAdditional revenue from existing customersUpsell/cross-sell effectiveness
Contraction revenueReduced revenue from downgradesValue perception issues
Churned revenueRevenue from lost customersProduct-market fit problems

Example -- January 2026 cohort, month 4:

ComponentAmountImpact
Starting MRR$10,980--
Retained (same plan)$9,200Base
Expansion (upgrades)$1,950+17.8%
Contraction (downgrades)($430)-3.9%
Churned($800)-7.3%
Ending MRR$10,920-0.5% net

This breakdown shows that while 7.3% of revenue churned, 17.8% expansion nearly offset it entirely. The problem to fix is not churn (which is in a reasonable range) but contraction -- $430 in downgrades suggests some customers are not seeing enough value to stay on their current plan.

Technique 3: Cohort Payback Analysis

Track not just revenue retention but profitability retention. A cohort that retains 100% of revenue but requires increasing support costs is less valuable than it appears.

MonthRevenueCost to ServeContributionCumulative Contribution
1$10,980$3,500$7,480$7,480
2$10,440$2,800$7,640$15,120
3$10,640$2,500$8,140$23,260
4$10,920$2,400$8,520$31,780

Notice that cost to serve decreases over time as customers need less support. This means the profitability of each cohort improves even if revenue is flat -- a dynamic that aggregate metrics completely miss.

Common Cohort Analysis Mistakes

Mistake 1: Using Too Short a Time Frame

Cohort analysis requires at least 6 months of data to show meaningful patterns. Early months are dominated by onboarding-related churn, which is a different problem than long-term retention. Wait for enough data before drawing conclusions.

Mistake 2: Ignoring Cohort Size Differences

A cohort of 10 customers showing 100% retention is less meaningful than a cohort of 100 showing 95%. Small cohorts create noisy data. Weight your analysis by cohort size or focus on cohorts with statistically significant sample sizes.

Mistake 3: Not Separating Voluntary and Involuntary Churn

Involuntary churn (expired credit cards, failed payments) is a payments problem, not a product problem. Separate it from voluntary churn to get a clear picture of product-driven retention.

Mistake 4: Looking Only at Revenue, Not Engagement

Revenue retention is a lagging indicator. By the time revenue drops, the customer has already disengaged. Build leading indicator cohorts based on product usage (logins, feature adoption, API calls) to predict revenue churn before it happens.

FAQ

How many months of data do I need to start cohort analysis?

You need at least 3 months to see any patterns and 6-12 months for reliable trends. Start building your cohort table now even if you only have 2 months of data -- the table will become more valuable every month.

Should I use monthly or weekly cohorts?

Monthly cohorts are standard for most SaaS businesses. Use weekly cohorts only if you have high volume (100+ signups per week) and need to measure the impact of rapid product changes or marketing campaigns.

What NRR should I target as a seed-stage company?

Target 90%+ gross revenue retention and 100%+ net revenue retention. If your NRR is below 90%, focus all product effort on retention before investing more in acquisition. Acquiring customers into a leaky bucket is the most expensive way to grow.

Sources

  • ChartMogul, "2025 SaaS Retention Benchmarks" (analysis of 2,000+ SaaS companies)
  • Bessemer Venture Partners, "State of the Cloud 2025"
  • ProfitWell, "2025 Retention Report"
  • SaaS Capital, "Annual Benchmarks Report 2025"
  • Lenny Rachitsky, "What is Good Retention?" (updated 2025)

Track cohort revenue retention automatically, get alerts when cohort performance drops, and benchmark against industry standards. Create your free culta.ai account and see the retention patterns your aggregate metrics are hiding.

T

Written by Team culta

The culta.ai team helps businesses track revenue, manage cash flow, and make smarter financial decisions across multiple entities.

Ready to get started?

Take control of your finances

Start free and use culta.ai to track revenue and make smarter financial decisions.