What net revenue retention measures
Net revenue retention (NRR) answers one question: if you stopped acquiring customers today, would the revenue from your existing customers grow or shrink over the next year? It takes the recurring revenue from a fixed group of customers at the start of a period and compares it with the recurring revenue from those same customers at the end, after upgrades, downgrades, and cancellations.
An NRR of 100% means expansion from existing customers exactly offset what was lost. Above 100%, the existing base grows on its own. Below 100%, the business has to win new customers just to stand still.
NRR is popular with investors and acquirers because it combines retention and expansion into one figure. It is also easy to calculate in a way that flatters the business. This guide covers the formula, a worked cohort example, how NRR differs from gross revenue retention (GRR), published benchmarks, and the reporting mistakes to avoid. To compute it from your own numbers, use the SaaS churn and retention calculator.
The net revenue retention formula
For a cohort of customers who were active at the start of the period:
NRR = (starting MRR + expansion − contraction − churned MRR) ÷ starting MRR
Equivalently:
NRR = ending MRR from the starting cohort ÷ starting MRR of that cohort
The second form is how SaaS Capital defines it in its 2025 retention benchmarks brief: MRR in December 2024 from customers who were already customers in December 2023, divided by total MRR in December 2023. That definition includes upsells, cross-sells, and price increases, and excludes any revenue from customers acquired during the year.
Two rules follow directly:
- New customers never enter NRR. Revenue from accounts acquired after the start date is excluded from both numerator and denominator.
- Churned customers stay in the cohort at zero. You do not drop them; their lost revenue is exactly what NRR is meant to capture.
Gross revenue retention: the companion metric
GRR uses the same cohort but ignores expansion:
GRR = (starting MRR − contraction − churned MRR) ÷ starting MRR
SaaS Capital calculates GRR with the same formula as NRR, excluding upsells, cross-sells, and price increases, so GRR cannot exceed 100%. ChartMogul's SaaS metrics cheat sheet describes GRR as the percentage of revenue retained excluding expansions.
Report both. NRR tells you whether the base grows; GRR tells you how much of it leaks out. A business with 110% NRR and 75% GRR is replacing heavy losses with upgrades from a smaller group of customers, which is a very different risk profile from 110% NRR with 95% GRR.
Worked example: a 12-month cohort
On October 1, 2025, a company has 300 customers paying $30,000 of MRR. Twelve months later, looking only at those 300 customers:
- 45 customers have canceled; they had been paying $3,600 of MRR;
- 30 customers downgraded, reducing MRR by $900;
- 60 customers upgraded or added seats, adding $6,300 of MRR;
- a price increase on one legacy plan added $600 of MRR across remaining customers.
Expansion, including the price increase, is $6,300 + $600 = $6,900.
NRR
($30,000 + $6,900 − $900 − $3,600) ÷ $30,000 = $32,400 ÷ $30,000 = 108%
GRR
($30,000 − $900 − $3,600) ÷ $30,000 = $25,500 ÷ $30,000 = 85%
Logo retention
(300 − 45) ÷ 300 = 85%
During the same year the company also won 120 new customers worth $9,000 of MRR, so its total MRR is now $41,400. Those new customers affect growth but not NRR. If someone calculated NRR as $41,400 ÷ $30,000 = 138%, they would be reporting total revenue growth under the wrong name.
Monthly NRR, trailing NRR, and annualizing
Many dashboards show monthly NRR. A monthly figure can be converted to an approximate annual rate by compounding:
Annual NRR ≈ (monthly NRR)^12
A monthly NRR of 100.5% compounds to about 1.005^12 ≈ 106.2%. A monthly NRR of 99% compounds to about 0.99^12 ≈ 88.6%. The small monthly differences become large annual ones.
The conversion is only approximate because it assumes every month behaves the same, and because a series of monthly cohorts is not identical to one annual cohort. When you report NRR to a buyer or investor, state whether it is:
- a single 12-month cohort (start date to end date);
- an average of several trailing 12-month cohorts; or
- monthly NRR compounded to an annual figure.
Stripe's Billing analytics documentation shows the cohort approach month by month: in its example, 100 subscribers start on $10 plans ($1,000 MRR); after 10 churn and 5 upgrade to $25 plans, Stripe counts $125 of expansion against $100 of churn and shows 102.5% revenue retention for the first month.
Public companies define NRR in their own ways
There is no accounting standard for NRR, and public companies publish different definitions. Snowflake's Form 10-K for the fiscal year ended January 31, 2026 calculates net revenue retention over a trailing two-year window: it takes customers under capacity contracts who used the platform in the first month of the first year, and divides their product revenue in the second year by their product revenue in the first year. It reported 125% as of January 31, 2026 and 126% as of January 31, 2025.
That is a consumption-revenue definition, not an MRR snapshot. It is a sensible definition for Snowflake's business, but it is not directly comparable with a subscription company's December-to-December MRR calculation. Whenever you compare NRR figures, read the definition first.
Net revenue retention benchmarks
Benchmarks vary widely by price point and customer type:
- SaaS Capital, 2025: among private B2B SaaS companies (charts exclude those under $1 million ARR), median NRR was 101% and median GRR 91%. Median NRR ranged from 98% for ACVs under $12,000 to 106% above $250,000. Bootstrapped companies reported 104% NRR and 92% GRR; equity-backed companies 101% and 90%.
- ChartMogul, 2023: its SaaS Retention Report, covering more than 2,100 SaaS businesses, found that only 2.7% of businesses with ARPA under $10 per month had NRR above 100%, compared with 41.1% of those with ARPA over $500 per month. Top-quartile NRR was 65.1% in the lowest band and 109.3% in the highest.
- ChartMogul, December 2025: the cheat sheet cites median NRR of 82% for B2B SaaS (97% top quartile), and around 48% to 49% for consumer and AI-native products.
The spread between SaaS Capital's 101% and ChartMogul's 82% B2B median reflects different samples, not a contradiction. For more on segment differences, see what is a good churn rate for SaaS.
Why NRR matters for growth and valuation
SaaS Capital's 2025 brief reports a strong relationship between NRR and growth. Across its sample, median growth was 24%. Companies with NRR of at least 110% grew faster than that median, while those below 100% grew slower. Median growth rose from 15% for companies with NRR below 90% to 50% for those above 130%, and moving from the 100%–110% band to 110%–120% was associated with 9 percentage points more growth. The brief found GRR much less correlated with growth, but describes it as a "table stakes" benchmark.
For a buyer, NRR shows whether today's revenue base is likely to be larger or smaller in a year before any new sales effort. That feeds directly into forecasts and multiples; see ARR multiples and test the effect with the SaaS valuation calculator.
Common NRR mistakes
Including new customers. This turns NRR into a growth rate. New-customer revenue belongs in growth, not retention.
Dropping churned customers from the cohort. Their starting MRR stays in the denominator and their ending MRR is zero.
Counting reactivations as expansion without disclosure. A customer who churned and came back is a policy choice; state it.
Ignoring FX. A currency move can raise or lower NRR without any customer behavior changing. Show it separately.
Reporting only NRR. Strong expansion from a few accounts can hide weak GRR and high concentration.
Comparing monthly NRR with annual benchmarks. Compound or measure the annual cohort directly.
Using payment totals instead of recurring revenue. RevenueBug's public trailing 30-day revenue figure is calculated from connected payment-provider data and can include annual prepayments and one-time sales, so it is not an NRR input by itself. The methodology explains what it covers; calculate NRR from customer-level subscription records. For how monthly recurring revenue itself is defined, see what is MRR.
Frequently asked questions
What is a good net revenue retention rate?
It depends on price point and customer type. SaaS Capital's 2025 median for private B2B SaaS was 101%, with higher figures at higher contract values. Low-priced and consumer subscriptions rarely exceed 100%.
What is the difference between NRR and GRR?
NRR includes expansion, so it can exceed 100%. GRR excludes expansion and is capped at 100%. GRR shows how much revenue you keep; NRR shows whether the existing base grows.
Can NRR be above 100% if customers are churning?
Yes. If upgrades, added seats, and price increases from remaining customers exceed the revenue lost to churn and downgrades, NRR is above 100% even with significant logo churn.
Is net revenue retention the same as negative churn?
They describe the same condition. NRR above 100% is equivalent to negative net revenue churn over the same period and cohort.
Do price increases count toward NRR?
In SaaS Capital's definition, yes: NRR includes upsells, cross-sells, and price increases. Disclose large price changes, because their effect may not repeat.
How often should I calculate NRR?
Track it monthly for early warning, but report annual cohort NRR for comparison with benchmarks and in sale or fundraising materials.