What is Dollar-Based Cohort Retention?
Dollar-Based Cohort Retention (DBR) is a SaaS analytics framework that tracks revenue retention from a specific cohort of customers (typically grouped by acquisition month or quarter) over time, measured in dollars rather than customer count. Unlike NRR/GRR which collapse all cohorts into a single period metric, DBR isolates the long-term behavior of each cohort — revealing whether customers acquired in different periods grow, shrink, or churn differently. DBR is the workhorse of investor cohort analysis and unit economics modeling.
How DBR is calculated
- Cohort definition: all customers acquired in a defined period (month, quarter, year).
- Vintage tracking: revenue from that cohort tracked monthly/quarterly thereafter — month 1, month 12, month 24, etc.
- DBR at month N = (Cohort revenue at month N) / (Cohort revenue at month 1) × 100
- Inclusion: upsell, contraction, churn all reflected; new customers excluded (those go into their own cohort).
What DBR reveals
- Land-and-expand effectiveness: DBR ≥120% at month 24 indicates strong expansion motion.
- Cohort decay: early-stage cohort DBR sharply below later-stage suggests product or onboarding maturation.
- Pricing or segmentation shifts: sudden cohort improvements reveal product-market fit milestones.
- LTV computation: DBR curve directly drives Customer Lifetime Value forecasts.
DBR benchmarks
- SMB SaaS at month 24: 70-100% DBR (high churn offset by moderate expansion).
- Mid-market SaaS: 100-130% DBR at month 24.
- Enterprise SaaS: 120-150%+ DBR at month 24.
- Best-in-class PLG/usage-based: 150-200%+ DBR (Snowflake, Twilio characteristic patterns).
Cohorts as the diligence ground truth
Dollar-based cohort tables are where diligence verifies every retention claim: take each acquisition cohort’s revenue at month zero and track it across months, and the company’s NRR/GRR statements either reproduce or they do not. Three reading disciplines: separate logo retention from dollar retention (few large expansions can mask broad logo churn), check whether early cohorts and recent cohorts behave alike (deteriorating recent cohorts is the growth-quality red flag), and confirm the revenue definition (booked, billed or recognised). When metrics feed earn-outs or covenants, write the cohort methodology into the document — measurement windows, upgrade/downgrade treatment, currency — because cohort arithmetic has too many degrees of freedom to leave implicit.
If this is on your desk
Templates and checklists are free in the Founder Academy; for a specific situation, book a 30-minute intro call.
Founder AcademyBook an intro call