Gross Retention Rate (GRR) Calculator
Compute Gross Revenue Retention (GRR) — the share of starting ARR retained without counting expansion. Always less than or equal to 100%.
GRR vs NRR — Two Sides of Retention
GRR (Gross Retention Rate): measures pure customer/revenue stickiness. Includes contraction and churn losses but EXCLUDES expansion. Bounded by 100%. NRR (Net Retention Rate): measures total revenue retention including expansion. Can exceed 100%.
Both matter. NRR > 100% with GRR = 95% is excellent — strong stickiness + strong expansion. NRR > 100% with GRR = 75% is alarming — heavy churn masked by upsells, fragile. Always report both — so this calculator now does. Enter an optional Expansion ARR and the result grid shows your GRR, your NRR, and the gap between them in percentage points, from one set of inputs. Read the gap this way: a gap under 10 pts means growth is coming from a genuinely sticky base; a gap above 40 pts means expansion is carrying a leaky base and slowing the upsell motion would stall growth. Source: Bessemer Cloud Index, OpenView Benchmarks 2026. Last updated: 2026-08-21.
GRR Benchmarks by Segment 2026
| Segment | Healthy GRR | Best-in-Class |
|---|---|---|
| Enterprise SaaS | 90-95% | 95%+ |
| Mid-Market SaaS | 85-90% | 92%+ |
| SMB SaaS | 75-85% | 88%+ |
| Self-serve / PLG | 65-80% | 85%+ |
Note: smaller customers churn more frequently — SMB GRR of 80% is normal; enterprise GRR of 80% would be alarming.
Why Investors Care About GRR
NRR is the headline metric, but sophisticated investors triangulate to GRR because it shows the 'floor' of customer loyalty before any expansion intervention. A company with 130% NRR and 95% GRR has organic growth even without aggressive upsell motion. A company with 130% NRR and 75% GRR is running an upsell-treadmill — slowing the upsell motion would tank growth.
How to Improve GRR
(1) Onboarding quality. Customers who reach activation/aha within 30 days churn at half the rate of those who don't. (2) Identify churn risk early. Customer health scores combining usage, NPS, support ticket frequency. (3) Make churn intentional. Require manual cancellation (not self-service). Add pause options. Survey leavers ruthlessly. (4) Address contraction. Often more recoverable than churn — find why customers are downgrading and what would make them re-upgrade.
Gross Retention Rate Examples: Real Public SaaS Numbers (2026)
Public SaaS filings disclosed to the SEC via EDGAR show the range of real-world GRR at scale. Example 1: Snowflake FY2024 10-K reported NRR 131% with disclosed GRR of roughly 95% — high stickiness plus consumption-based expansion. Example 2: Datadog reported NRR "above 130%" for years while GRR held ~93-95% — a healthy gap because expansion is genuine, not a mask. Example 3: A hypothetical SMB SaaS starts a quarter at $4,000,000 ARR, loses $80,000 to churn and $40,000 to contraction — GRR = ($4,000,000 − $120,000) / $4,000,000 = 97.0%, best-in-class SMB. Example 4: A PLG tool starts at $2,000,000, loses $300,000 to churn, $100,000 to contraction — GRR = 80.0%, healthy for self-serve but concerning if positioned as enterprise. Compare against Bessemer's State of the Cloud benchmark tables to place your number in context. You do not have to retype any of these: the three “Load a worked example” buttons in the calculator above fill in the SMB (97.0%), PLG (80.0%) and enterprise (95.0%) scenarios and compute them in one click, so you can see how each input moves the number before swapping in your own ARR. Updated 2026-08-12.
Frequently Asked Questions
What is the difference between GRR and NRR?
GRR excludes expansion (capped at 100%); NRR includes expansion (can exceed 100%). GRR measures pure stickiness; NRR measures total revenue retention including upsell. Both matter \u2014 strong companies report both.
What is a good GRR for SaaS?
Enterprise: 90%+ healthy, 95%+ best-in-class. Mid-market: 85-92%. SMB: 75-88%. Self-serve / PLG: 65-85%. Smaller customers churn more, so context matters \u2014 don't compare an enterprise SaaS to a PLG product directly.
Can GRR be over 100%?
No \u2014 by definition, GRR is bounded by 100%. GRR measures retained portion of starting ARR; you can't retain more than you started with (only NRR exceeds 100% via expansion ARR).
Why does my GRR exceed my NRR?
Mathematical impossibility. GRR (no expansion) is always \u2264 NRR (includes expansion). If you're computing higher GRR than NRR, you have a calculation error \u2014 most likely double-counting churn or treating contraction as expansion.
How is contraction different from churn?
Churn = customer fully cancels (ARR drops to zero). Contraction = customer reduces (downgrades to lower plan, reduces seat count, drops products) but stays. Both count against GRR. Track separately \u2014 contraction is often more recoverable than churn.
Is GRR reported in public SaaS company filings?
Sometimes. Companies that disclose 'Net Dollar Retention' or 'Net Revenue Retention' often don't separately disclose GRR. Some investor day presentations split it out. Public 10-K filings vary. Bessemer Cloud Index and OpenView Benchmarks aggregate disclosed data.
Can you give a gross retention rate example calculation?
Yes. Start ARR $4,000,000. During the quarter, $80,000 churns and $40,000 contracts. GRR = ($4,000,000 − $80,000 − $40,000) / $4,000,000 = $3,880,000 / $4,000,000 = 97.0%. Expansion is IGNORED for GRR — even if the same cohort added $200,000 in upsell, GRR stays 97.0% (that expansion only lifts NRR). Enter these numbers in the calculator above to verify.
What is a real-world gross retention rate example from public SaaS?
Snowflake's FY2024 10-K reported NRR ~131% with disclosed GRR around 95% (consumption-based expansion, low churn). Datadog held NRR above 130% multiple years while GRR stayed ~93-95%. CrowdStrike disclosed GRR ~98% in investor days. These are best-in-class enterprise numbers — SMB and PLG SaaS typically sit 15-25 points lower. Source: SEC EDGAR 10-K filings, Bessemer Cloud Index.