GRR Shows the Floor, And the Floor Is What Keeps You Alive
If net revenue retention includes the upside, GRR shows the floor. GRR strips away the optimism of expansion revenue and shows you the raw retention picture: how much of your existing recurring revenue survives contractions and churn each year. A company with 110% NRR and 80% GRR is masking a serious retention problem with strong upsells. See the revenue operations guide for how retention metrics fit into the broader RevOps framework. Remove the expansion, and the base is eroding fast.Median GRR for B2B SaaS is 88% in Benchmarkit's 2025 benchmarks, down from 90% over three years. That means the median company loses about 12% of its existing revenue each year before any expansion.
GRR Benchmarks
Published benchmarks cluster between the high 80s and 90%, and account size moves the number more than anything else. These are the figures we could verify at the source:| Source | Benchmark | Scope |
|---|---|---|
| Benchmarkit, 2025 | Median GRR of 88%, down from 90% over three years | B2B SaaS survey |
| High Alpha, 2025 | Keeping nine of ten customers is the norm at every ARR band | 800+ respondents |
| ChartMogul | Its top mark is anything over 86%, at any stage | Subscription businesses |
| ChartMogul | Top quartile reaches about 88% where average revenue per account is over $500 a month, and 60% to 70% where it is under $50 | By account size |
How Do You Calculate GRR from a Revenue Waterfall?
Build a monthly ARR waterfall that reconciles, and GRR falls out of it. ORM builds this bridge for every customer. Each month starts with the prior month's ending ARR, then walks through every change:| Line | What it captures | Hypothetical example |
|---|---|---|
| Beginning ARR | Ending ARR from the prior period | $10,000,000 |
| Churned customer ARR | Customers who left | -$500,000 |
| Churned product ARR | Products a customer dropped | -$150,000 |
| Product decrease ARR | Fewer seats or a smaller plan | -$250,000 |
| New product ARR | Products added by existing customers | +$400,000 |
| Increased product ARR | More seats or a bigger plan | +$600,000 |
| New customer ARR | New logos | +$1,200,000 |
| Ending ARR | Beginning ARR plus every line above | $11,300,000 |
In the example, GRR is ($10,000,000 - $900,000) / $10,000,000 = 91.0%. New customer ARR stays out, because GRR measures only the base you started with. The same waterfall gives you net revenue retention: add the two expansion lines back and NRR is 101.0%.
The reason to build it monthly is reconciliation. When every month's beginning ARR equals the prior month's ending ARR, nobody argues about the number, and you can see exactly which line moved.
Why Is GRR Easier to Forecast Than New Business?
Because the amount and the date are mostly known. A renewal is usually the same as this year's contract or carries a price lift, and it closes around the renewal date, typically within about 30 days either side. New business depends on when a deal last changed stage, which is far less certain. Forecast renewals on their own track, and blending them into a new-business accuracy figure makes that figure look better than it is.Why Finance Teams Love GRR
GRR is the metric finance uses to stress-test revenue plans because it answers the worst-case question: "What happens if we stop selling entirely?" If GRR is 95%, the business retains $95 of every $100 in existing revenue, a strong foundation even without new logos. If GRR is 80%, the business loses $20 of every $100, meaning you need significant new and expansion revenue just to stay flat.That stress test is critical during board planning and investor diligence. For the full list of metrics that matter in these conversations, see RevOps KPIs. A company with strong NRR but weak GRR is a company that depends on its expansion motion to survive. If expansion slows for any reason, team changes, market shifts, product maturity, the floor drops quickly.
How to Improve GRR
GRR improves through two levers: reducing churn and reducing contraction. Reducing churn means improving product value, deepening integrations, and building switching costs into the customer relationship. Reducing contraction means identifying at-risk accounts before they downgrade, through usage monitoring, health scores, and proactive outreach. Track GRR by cohort and segment to find where the base is weakest. The fix is always specific to the segment losing the most revenue, not a company-wide initiative that spreads effort thin. Pair GRR with NRR to see both the floor and the ceiling, together they tell the full retention story.Frequently Asked Questions
How do you calculate gross revenue retention?
Take beginning ARR, subtract churned customer ARR, churned product ARR and product decreases, then divide by beginning ARR. Leave out expansion and new customers. GRR can never be above 100%.
What is the difference between GRR and NRR?
GRR counts only what you keep, so it can never exceed 100%. NRR also adds expansion from existing customers, so it can run above 100%. Neither includes new customers.
What is a healthy GRR for B2B SaaS?
Benchmarkit's 2025 survey puts median GRR at 88%, down from 90% over three years. High Alpha's 2025 report says keeping nine of ten customers is now the norm at every ARR band. ChartMogul sets its top mark at anything over 86%, at any stage.
Why does GRR matter if NRR is above 100%?
NRR includes the upside from expansion revenue. GRR shows the floor by removing the optimism of expansion and showing the raw retention picture, how well you keep what you already have.
Why do finance teams prefer GRR for stress testing?
Because it removes the optimism of expansion. GRR answers the worst-case question: how much of today's recurring revenue survives if you stop selling? A strong NRR built on a weak GRR depends on expansion to hide churn.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like gross revenue retention (grr) into prescriptive action for your team.
Schedule a Demo