GRR and NRR

Two retention metrics dominate SaaS board decks. Gross revenue retention (GRR) — sometimes gross dollar retention — asks: of the ARR you had from a starting base a year ago, how much remains before counting expansion? Net revenue retention (NRR) — net dollar retention — asks the same question including expansion.

NRR can exceed 100% when upsells outrun downgrades and cancels. David Skok called that negative churn.

The big idea

GRR is "how much of last year's book did we keep?" NRR is "how much is that same book worth now?" Expansion is the difference.

The analogy of a leaking bucket helps: GRR measures the holes. NRR measures holes plus the rain you collected from people already in the bucket. New logos are a different bucket. Keep them out of this ratio.

Walking through it

For a starting base of customers (exclude brand-new logos):

GRR = (Starting ARR − Contraction − Churn) / Starting ARR
NRR = (Starting ARR + Expansion − Contraction − Churn) / Starting ARR

Starting ARR from the retained base: $10M. Over the next 12 months that same base produces +$2M expansion, −$0.5M contraction, −$1M churn.

GRR = (10.0 − 0.5 − 1.0) / 10.0 = 8.5 / 10.0 = 85%
NRR = (10.0 + 2.0 − 0.5 − 1.0) / 10.0 = 10.5 / 10.0 = 105%

You lost 15% of starting dollars on a gross basis, but expansion more than filled the hole. The same customers are worth 5% more than a year ago. That is the engine behind "growth even if we stopped acquiring logos."

Public cloud companies often print NRR in the 110–130%+ range; best-in-class land north of 130%. GRR in the 85–95% band is common. Below about 80% annual GRR usually means the product or segment fit is weak no matter how good new sales look.

These numbers are cleanest on cohorts: everyone who was a customer on 1 Jan, what they pay next 1 Jan. Company-level ARR bridges approximate the same idea. Either is fine if you are consistent. Switching definitions between fundraises is not.

Rough mid-2020s intuition, not law:

SMB can still be excellent if CAC is cheap enough that high churn still yields good LTV/CAC, or if ACV and NRR climb over time.

Common mistakes

• Putting new logos in NRR. That is growth, not retention.
• Celebrating NRR while GRR is 70%. Expansion is hiding a product problem.
• Changing the formula after a weak year (exclude M&A, change the cohort rule).
• Treating 105% NRR from price hikes as the same as seat growth. Fragile vs durable.

Check yourself

$10M start, +$2M expansion, −$0.5M contraction, −$1M churn. GRR and NRR?

GRR 85%, NRR 105%.

Why is GRR capped at 100%?

It ignores expansion. You cannot keep more than you started with on a gross basis.

Key takeaways

• GRR ignores expansion and is capped at 100%. NRR includes expansion and can exceed 100%.
• Worked pattern: $10M start, +$2M, −$0.5M, −$1M → GRR 85%, NRR 105%.
• Below ~80% annual GRR, new sales are filling a leaky bucket.
• Pick a cohort definition and do not quietly change it.

Further learning

SaaS Metrics 2.0 (David Skok)“If you cannot measure it, you cannot improve it” – Lord Kelvin This article is a comprehensive and detailed look at the key metrics that are needed to understand and optimize a SaaS business. It is a completely updat…For Entrepreneurs

16 Startup Metrics (a16z)We have the privilege of meeting with thousands of entrepreneurs every year, and in the course of those discussions are presented with all kinds of numbers, measures, and metrics that illustrate the promise and health…Andreessen Horowitz

Next: SaaS quick ratio.