Net Revenue Retention
The cohort retention metric that shows whether existing recurring revenue expands, shrinks, or quietly masks churn.
A SaaS company can add new annual recurring revenue every quarter and still have a sick customer base underneath it. New logos can hide churn, expansions can hide contractions, and a growing topline can conceal the fact that last year’s customers are worth less than they used to be. Net revenue retention (NRR), also called net dollar retention, cuts away the new-logo story and asks a narrower question: what happened to the revenue from the customers who were already here?
What It Is
Net revenue retention measures the recurring revenue kept from an existing customer cohort after churn, contraction, expansion, cross-sell, upsell, usage growth, and price changes are netted together. The clean version is cohort-based. Start with the recurring revenue from customers present at the beginning of the period, then compare it with recurring revenue from that same cohort at the end. New customers are excluded.
NRR = (starting_recurring_revenue - churn - contraction + expansion) / starting_recurring_revenue
A company that starts the year with $1M of recurring revenue from an existing cohort, loses $100K to churn and contraction, and gains $250K from expansion ends at $1.15M from the same cohort. Its NRR is 115%. That means the existing base grew before any new customer was counted.
The metric differs from gross revenue retention (GRR). GRR measures how much recurring revenue stays before expansion is added back, so it can’t exceed 100%. NRR can exceed 100% because expansion can more than offset losses. Logo retention is different again: it counts accounts, not dollars. A company can keep most logos while losing revenue if large customers downsize, or lose many small logos while revenue grows because a few large accounts expand.
Why It Matters
NRR tells a founder whether growth compounds or leaks. A business with NRR above 100% starts each period with more revenue from the same customers than it had before. That makes new sales additive. A business with NRR below 100% has to refill the bucket before it grows, so a large share of acquisition spend is replacing revenue that left.
The investor reads NRR as revenue quality. Strong retention means the company doesn’t have to resell the same revenue every year, and expansion inside the base suggests the product is becoming more useful after the first sale. Weak retention forces a harder question: is the company growing because customers deepen their commitment, or because sales keeps finding replacements fast enough to keep the chart rising?
Startup talent can read the same number as a company-health signal. High NRR is not proof of a great company, but low NRR is hard to ignore in a recurring-revenue business. It means the product may be useful enough to buy once but not useful enough to keep, expand, or embed in the customer’s operating routine. Equity in that company carries a different risk than equity in one where the existing base grows on its own.
What NRR gives the practitioner is a way to separate acquisition from durability. New ARR says the company can sell. NRR says whether customers stay and grow after sales leaves the room.
How to Recognize It
A useful NRR figure names the cohort, the period, and the revenue type. Annual recurring revenue (ARR) and monthly recurring revenue (MRR) should not be mixed. A trailing twelve-month NRR is usually more reliable than a single-quarter figure because expansion and renewal timing can distort short windows.
The 2025 private B2B SaaS benchmark frame puts median NRR near 101%, which is barely above the line where expansion offsets churn and contraction. In that market, 100% is not an excellence threshold. It is the point where the existing base has stopped shrinking.
| NRR range | Common reading |
|---|---|
| Below 90% | The base is shrinking materially; new sales are replacing lost revenue. |
| 90-100% | Revenue is mostly retained but not expanding enough to offset all losses. |
| 100-110% | The base is expanding modestly; growth has a retention tailwind. |
| Above 110% | Existing accounts are expanding strongly, usually from upsell, seat growth, or usage growth. |
The table travels poorly outside recurring-revenue software. A seat-based collaboration tool, a usage-priced infrastructure product, and a high-touch enterprise application can all post different healthy ranges because expansion mechanics differ. Read the range as a diagnostic prompt, not a universal grade.
Never read NRR without GRR beside it. A company with 115% NRR and 95% GRR is expanding a healthy base. A company with 115% NRR and 65% GRR may be losing a wide set of customers while a few large accounts grow fast enough to hide the damage. The first business compounds. The second is fragile, even though the headline NRR is identical.
How It Plays Out
The clean case is an enterprise workflow product that starts the year with $10M ARR from existing customers. A few small customers churn, some accounts reduce seats, and the losses total $700K. But the product is now embedded in more teams inside the largest accounts, producing $1.8M of expansion. End-of-year cohort revenue is $11.1M, so NRR is 111% and GRR is 93%. That is the pattern investors like: losses exist, but expansion from retained customers more than covers them.
The trap case can post the same NRR. Another company starts with the same $10M cohort, loses $3.5M to churn and contraction, and expands a handful of large accounts by $4.6M. NRR is still 111%, but GRR is only 65%. The company has a few accounts that love it and a broad base that is leaving. If the board sees only NRR, the business looks durable. If it sees GRR beside NRR, the shape is obvious: the base is hollowing out under the winners.
The product-led growth version appears after the free-to-paid motion begins working. A self-serve tool converts thousands of small teams, then watches successful accounts add seats, upgrade plans, and invite adjacent teams. New-logo acquisition may be cheap, but the stronger signal is that accounts keep getting larger after the first purchase. NRR is the proof that the product is not only acquiring users; it’s expanding inside the customer.
Consequences
Treating NRR as a core metric changes which growth stories a company can defend.
Benefits. NRR makes expansion and contraction visible in the same number, so a founder can see whether the customer base is compounding before adding new logos. It gives investors a compact diligence test for recurring revenue quality. It also ties directly into unit economics: stronger retention raises lifetime value, supports a healthier CAC/LTV ratio, and makes customer-acquisition spend easier to defend.
Liabilities. NRR is easy to overstate by using a favorable cohort, a short period, or a blended customer base that hides segment differences. It can also flatter a company with weak gross retention when a few large accounts expand enough to offset broad churn, the same measurement trap behind Vibe Revenue. And it says little about cash timing. A customer base can expand over a year while acquisition still pays back too slowly, which is why NRR travels with CAC Payback Period, not instead of it.
NRR answers one question well: does revenue from existing customers grow after the first sale? It doesn’t answer whether the company can win customers cheaply, collect cash soon enough, or defend the product against a better alternative.
Related Articles
Sources
- SaaS Metrics Standards Board, Net Revenue Retention — formalizes the cohort-based definition and formula family used here.
- ChartMogul, SaaS Retention: The New Normal — frames 100%+ NRR as a threshold associated with faster SaaS growth and separates NRR from gross retention.
- SaaS Capital, Private SaaS Company Growth Rate Benchmarks and What Is a Good Retention Rate for a Private SaaS Company? — provide the 2025-2026 private-SaaS benchmark frame for retention and growth.
- Benchmarkit, 2025 B2B SaaS Benchmarks — reports the 2025 private B2B SaaS median NRR used as the current-market reference point.