Gross Revenue Retention (GRR)

GRR tells you how much of your existing recurring revenue you kept, ignoring any expansion. No upsells, no cross-sells, no add-ons. Just: of the revenue you had at the start of the period, how much is still there? It's the purest measure of product stickiness you've got.

The formula

GRR = (Starting MRR − Contraction MRR − Churned MRR) / Starting MRR × 100

Notice what's missing: expansion MRR. That's the whole point. GRR can never exceed 100%. If it does, something's wrong with your calculation.

Example: you start the month with $100,000 in MRR. Downgrades cost $1,500 and churned accounts lose $3,000. Your GRR is 95.5%, regardless of whether upgrades added $10,000 or $0.

Benchmarks

ChurnZero's data puts median B2B SaaS GRR at around 90%. Above 95% is strong. Below 85% suggests your product has a retention problem that expansion revenue is papering over. Enterprise SaaS with long contracts and high switching costs typically sits at 95%+. SMB-focused products with monthly billing tend to land lower, around 85-92%.

Put simply, GRR is the metric that can't lie to you.

The gap between your GRR and your NRR reveals how dependent you are on upselling to compensate for losses. A company with 85% GRR and 115% NRR is growing, but it's working hard to replace the revenue walking out the door. A company with 97% GRR and 110% NRR has a much stronger foundation.

Why GRR matters alongside NRR

Investors increasingly ask for both. NRR can mask churn when expansion is strong. GRR can't. If your GRR is declining quarter over quarter, it means your core product is becoming less sticky, even if NRR looks healthy. For the full comparison of when to use each, see NRR vs GRR.

One practical lever we've seen work: a portion of the churn dragging GRR down is involuntary. Customers whose payments failed, not customers who chose to leave. Recovering those through dunning directly improves GRR without touching pricing, product, or customer success.