What is GRR?
GRR — gross revenue retention — is the cleanest measure of how well you hold on to the revenue you have already sold. It is the floor under NRR.
Quick answer — What does GRR stand for? GRR stands for gross revenue retention: the percentage of recurring revenue kept from existing customers after churn and downgrades, excluding expansion.
What does GRR stand for?
GRR stands for gross revenue retention. It measures the share of recurring revenue you keep from existing customers over a period, counting only losses (churn and contraction) and ignoring any expansion. Because expansion is excluded, GRR can never go above 100%.
Simple definition: GRR asks: of the revenue we started with, how much is still here, ignoring upsells?
The GRR formula
Using the same example as the NRR guide: a cohort starts at $1,000,000, loses $50,000 to downgrades and $100,000 to cancellations. GRR is 85%, even though NRR was above 100% thanks to $200,000 of expansion.
That gap is the point. Strong NRR can hide a weak GRR if a few large customers expand while many others leave.
Worked example with segments
Using the same illustrative cohort as the NRR guide: $1,000,000 starting MRR.
| Segment | Starting MRR | Contraction | Churn | Retained | GRR |
|---|---|---|---|---|---|
| Enterprise | $500,000 | −$10,000 | −$20,000 | $470,000 | 94% |
| Mid-market | $300,000 | −$20,000 | −$40,000 | $240,000 | 80% |
| SMB | $200,000 | −$20,000 | −$40,000 | $140,000 | 70% |
| Total | $1,000,000 | −$50,000 | −$100,000 | $850,000 | 85% |
The blended 85% hides a big spread: enterprise is stable, SMB is leaking. Because GRR ignores expansion, it shows the true stickiness of each group. Investors often look at GRR first to judge product-market fit before they look at NRR.
Common GRR mistakes
- Letting expansion offset losses. If expansion is included anywhere in the calculation, it is no longer GRR.
- Capping at the wrong level. A customer who expands 50% and then churns half should not cancel out.
- Ignoring downgrades. Contraction counts as lost revenue.
- Using inconsistent cohorts. Compare like with like across periods.
- Mixing monthly and annual contracts. Churn on a monthly plan is visible every month, so annualize carefully.
GRR across contract types
| Contract model | How GRR behaves |
|---|---|
| Annual contracts | Losses show up in lumps at renewal; measure on cohorts renewing in the period |
| Multi-year contracts | Churn is delayed; GRR can look strong until large cohorts hit renewal. Watch leading indicators |
| Monthly subscriptions | Churn is continuous; GRR moves with every cancellation and downgrade |
| Usage-based | Contraction can happen without cancellation. See usage-based pricing |
GRR vs NRR
| GRR | NRR | |
|---|---|---|
| Includes expansion | No | Yes |
| Maximum value | 100% | Unlimited |
| Best for judging | Product stickiness and base health | Overall growth from the installed base |
| Most affected by | Churn and downgrades | Churn, downgrades, and upsell |
How CS teams improve GRR
- Fix onboarding so customers reach value quickly. See time to value.
- Watch early-warning signals with a health score.
- Start renewals early with a clear renewal playbook.
- Understand why customers churn and fix the top root causes.
- Prevent downgrades by driving adoption of the features customers pay for.
Reading the number
Benchmarks vary widely by segment. Enterprise-heavy businesses usually sit higher than SMB-heavy ones because large customers cancel less often. Treat any public benchmark as a rough guide and compare against companies with similar customer size and contract length. The related churn rate is the mirror image of GRR.
Frequently asked questions
Benchmark your team against real CSM pay
Check an offer or a comp plan against crowdsourced CSM compensation data.