NRR Calculator — Net Revenue Retention

Q: How do I annualise a monthly NRR figure?

Raise it to the twelfth power: annual NRR = (monthly NRR)^12. A monthly NRR of 102% compounds to about 127% annually. Multiplying the monthly surplus by twelve overstates it — retention compounds like interest.

Net revenue retention and gross revenue retention from the same four cohort inputs — the pair every diligence deck asks for.

What this calculator measures

NRR — net revenue retention — asks one question: if you signed no new customers at all, what would happen to revenue? It tracks a fixed cohort (everyone who was a customer at the period start) and nets their expansion against their churn and downgrades. Above 100%, the installed base grows by itself. This calculator also returns GRR from the same inputs, because NRR without GRR is only half the truth.

The formulas

NRR = (starting MRR + expansion − contraction − churn) ÷ starting MRR × 100
GRR = (starting MRR − contraction − churn) ÷ starting MRR × 100
Annualised NRR = (monthly NRR)¹²

The cohort discipline is the whole metric: revenue from customers acquired during the period never enters either side of the fraction. NRR and NDR are the same calculation under different names.

Worked example

A cohort starts the month at $100,000 MRR. Existing customers add $6,000 in upgrades, downgrade by $1,500 and cancel $2,500. Ending cohort MRR = $102,000, so NRR = 102%. GRR = ($100,000 − $4,000) ÷ $100,000 = 96%. Held for a year, 102% monthly compounds to roughly 127% annual NRR — the base grows 27% with zero new sales.

What good looks like

Segment sets the bar. Enterprise SaaS: 110–130% annual NRR is strong, with the best public names historically printing 120%+. Mid-market: 100–115%. SMB: 90–105% is normal — small businesses fail and downgrade at rates no success team can fully offset. For GRR: above 90% is solid, above 95% excellent, and below 80% signals a product problem expansion is papering over. Full ranges on the benchmarks page.

Common mistakes

Letting new logos leak in. Any new-customer revenue in the numerator inflates NRR. The cohort is frozen at the period start.
Quoting NRR without GRR. Two whale accounts expanding can hold NRR at 105% while a third of customers churn. GRR exposes it.
Annualising by adding. 102% monthly is not 124% annual — it is 102%¹² ≈ 127%. Compounding works both directions: 99% monthly is 89% annual.
Comparing NRRs with different windows. Monthly, quarterly and TTM NRR are different numbers for the same business. State the window; check it when reading others’ claims.

FAQ

Do new customers count in NRR?

No. NRR measures only the cohort of customers you had at the period start. Revenue from logos acquired during the period is excluded from both numerator and denominator — NRR isolates how the existing base behaves, separate from new sales.

What is the difference between NRR and GRR?

NRR includes expansion revenue and can exceed 100%; GRR excludes it and is capped at 100%. NRR shows whether the base grows on its own; GRR shows how leaky the bucket is. Heavy expansion from a few accounts can hold NRR above 100% while GRR reveals serious churn underneath.

How do I annualise a monthly NRR figure?

Raise it to the twelfth power: annual NRR = (monthly NRR)¹². A monthly NRR of 102% compounds to about 127% annually. Multiplying the monthly surplus by twelve overstates it — retention compounds like interest.