Gross revenue retention (GRR) and net revenue retention (NRR) are the two metrics that, taken together, reveal whether your SaaS business is retaining real value or masking churn with expansion revenue. Most operators track one or the other; the ones who track both, and specifically measure the gap between them, gain a diagnostic tool that exposes structural problems before they show up in quarterly revenue misses.
GRR measures the percentage of recurring revenue retained from your existing customer base, excluding any expansion (upsells, cross-sells, seat additions). It can never exceed 100%. NRR includes that expansion revenue, so it can, and in healthy SaaS companies often does, exceed 100%. The spread between these two numbers is where the real story lives. A company with 88% GRR and 120% NRR looks great on the NRR line, but that 32-point gap means the business is losing nearly 12% of its base revenue every year and depending entirely on account expansion to cover the hole. That is not retention health. That is a treadmill.
To understand why the gap matters more than either number in isolation, think about what each metric actually rewards. GRR rewards low churn and low contraction. If your GRR is 95% or above, customers are staying and they are staying at roughly the same contract value. That is a sign your core product delivers what was promised during the sales cycle, your onboarding works, and your pricing holds up under scrutiny after the first renewal. NRR rewards expansion, which can come from genuine product-led growth (customers adopt more of the platform because it solves more problems) or from aggressive upselling that may or may not correlate with customer satisfaction. A high NRR with a mediocre GRR often means your sales team is very good at growing accounts that are quietly unhappy with the base product.
The most useful way to think about the gap is as a ratio of dependency. How much of your growth story depends on selling more to existing customers versus simply keeping them? Both matter, but they represent fundamentally different operational capabilities, and they break for different reasons.
I have seen mid-market SaaS companies celebrate 115% NRR at board meetings while ignoring a GRR that was sliding from 92% to 87% over three quarters. The NRR number held because a handful of large accounts expanded significantly, but the base was eroding. Within two more quarters, expansion could not keep pace. The accounts that would expand already had, and the shrinking base meant fewer accounts left to grow. This is the treadmill failure mode, and it is predictable if you watch the gap.
Calculating both metrics correctly is where many finance and ops teams stumble. The formulas are straightforward in theory but messy in practice.
| Metric | Formula | Cap |
|---|---|---|
| GRR | (Starting MRR − Churn − Contraction) ÷ Starting MRR | 100% |
| NRR | (Starting MRR − Churn − Contraction + Expansion) ÷ Starting MRR | None |
| Gap | NRR − GRR | Indicates expansion dependency |
The practical complications come from how you classify revenue movements. A customer who downgrades from a Pro plan to a Starter plan and then adds three seats: is that contraction, expansion, or both? Most billing systems either cannot track these transitions granularly or require manual tagging that drifts out of accuracy within weeks. If your billing infrastructure was not designed to capture plan changes, seat-level adjustments, and usage fluctuations as discrete events, your GRR and NRR calculations will be approximations at best. This is one of the reasons operators at companies with hybrid or usage-based pricing models end up rebuilding their billing layer earlier than they expected; the reporting demands outgrow the system long before the transaction volume does. Platforms like Market Rithm's Account Console were built with this kind of event-level revenue tracking in mind, specifically because accurate retention math depends on it.
Once you have clean numbers, the diagnostic work begins. A gap of 5 to 10 points between GRR and NRR is healthy in most B2B SaaS contexts. It means you have some expansion motion (good) but you are not relying on it to paper over attrition (also good). A gap of 15 to 20 points is a yellow flag. Above 20 points, you should be running a root-cause analysis on churn and contraction before your next board meeting.
The root causes that drive GRR down tend to cluster in a few areas. Poor onboarding is the most common. Customers who never reach first value in the first 30 days churn at rates two to three times higher than those who do. Pricing misalignment is the second: if customers feel the product is overpriced relative to the value they actually extract (as opposed to the value your marketing claims), they downgrade or leave. Product gaps rank third, usually manifesting as feature requests that accumulate without resolution until the customer finds an alternative. Support responsiveness is fourth, and it compounds the other three; a customer who is struggling with onboarding, unsure about value, or waiting on a feature fix will tolerate all of it longer if support is fast and genuinely helpful.
This is where customer success tooling earns its budget. The companies with the strongest GRR numbers (95%+ in B2B SaaS) almost always have some form of proactive health monitoring, whether that is a customer success platform, a set of internal dashboards, or, increasingly, AI-driven systems that flag at-risk accounts based on engagement patterns rather than waiting for a cancellation request. The shift toward AI-powered customer communication is particularly relevant for mid-market companies that cannot afford a dedicated CSM for every account. Per-conversation pricing models, like the approach Aigotchu takes, make it possible to deploy proactive outreach at scale without the fixed cost of an enterprise CS platform.
Expansion revenue, the driver of the NRR side of the gap, has its own set of dependencies. Product-led expansion (where customers adopt more features, add users, or increase usage organically) is the most durable form because it correlates with deepening product value. Sales-led expansion (where account managers proactively sell additional modules or upgrades) can be effective but tends to plateau once the obvious opportunities are captured. The healthiest companies I have worked with generate 60% or more of their expansion through product-led motions and use sales-led expansion for large strategic accounts where the deal complexity justifies human involvement.
Usage-based and hybrid pricing models tend to produce higher NRR naturally, because revenue grows with consumption without requiring a discrete upsell event. But they can also introduce GRR volatility if usage fluctuates seasonally or if customers can scale down without friction. If you are on a usage-based model, watch your GRR on a monthly basis, not just annually. Annual GRR can smooth over seasonal dips that, if they become structural, will not show up until it is too late to intervene.
Cohort analysis adds another layer of usefulness to the GRR/NRR gap. Measuring the gap for each customer cohort (by signup quarter, by plan tier, by acquisition channel) often reveals that your aggregate numbers are hiding wildly different retention profiles. I have seen companies where enterprise accounts had a 97% GRR and a 5-point gap, while SMB accounts had an 82% GRR and a 25-point gap. The blended number looked acceptable. The SMB segment was in freefall. You cannot fix what you cannot see, and cohort-level GRR/NRR splits are the fastest way to see it.
One more thing operators often overlook: the gap tells you something about your forecasting reliability. A company with a small, stable gap between GRR and NRR can forecast revenue with high confidence because the base is solid and expansion is predictable. A company with a large or widening gap is essentially making a bet every quarter that expansion will outpace attrition, and that bet gets harder to win as the denominator shrinks. If you are presenting revenue projections to investors, lenders, or your own exec team, showing the gap trend alongside the absolute numbers demonstrates operational maturity and gives everyone a clearer picture of where growth is actually coming from.
The takeaway for operators is straightforward: track both metrics, calculate the gap, trend it quarterly, and break it out by cohort. If the gap is widening, prioritize GRR improvements (onboarding, support responsiveness, pricing alignment) before doubling down on expansion. If GRR is strong and the gap is narrow, you have earned the right to invest in expansion motions. Start with the foundation. The rest follows.
What is the difference between GRR and NRR?
GRR (gross revenue retention) measures the percentage of recurring revenue retained from existing customers, excluding any expansion revenue. It caps at 100%. NRR (net revenue retention) includes expansion revenue from upsells, cross-sells, and seat additions, so it can exceed 100%. GRR tells you how well you keep what you have; NRR tells you how well you grow what you keep.
What is a good GRR for a B2B SaaS company?
Most healthy B2B SaaS companies target a GRR of 90% or higher. Enterprise-focused companies often achieve 95%+. If your GRR is below 85%, you likely have a systemic problem in onboarding, pricing, or product-market fit that expansion revenue alone cannot sustainably offset.
How large should the gap between GRR and NRR be?
A gap of 5 to 10 percentage points is typical and healthy in B2B SaaS. It indicates meaningful expansion without over-reliance on it. Gaps above 20 points signal that your business is compensating for high churn or contraction with aggressive account expansion, which becomes increasingly difficult to sustain over time.
Why might NRR look healthy while the business is struggling?
NRR can mask churn when a small number of large accounts expand significantly, covering the revenue lost from many smaller accounts that downgrade or cancel. This is why GRR is the better indicator of baseline retention health. Tracking both metrics, and the gap between them, prevents a false sense of security from a strong NRR number alone.
How often should operators review GRR and NRR?
Monthly tracking is ideal, especially for companies with usage-based pricing where revenue can fluctuate with consumption patterns. At minimum, review both metrics quarterly, segmented by customer cohort (plan tier, acquisition channel, signup date). Annual reviews smooth over seasonal and structural shifts that need earlier intervention.
