For a founder-CEO navigating the $10M to $50M revenue canyon, Net Revenue Retention (NRR) is the seductive metric. It tells a story of growth, compounding value, and product-market fit. You look at your dashboard, see 110% NRR, and tell the board, "We're growing even if we don't sell a single new logo."
But for 40% of the companies we analyze at Human Renaissance, that 110% is a lie. It is masking a rot in the core business that will inevitably stall growth and crush your exit multiple.
The problem is the expansion mask. In B2B SaaS and tech services, it is entirely possible to have an NRR of 115% while hemorrhaging 15% of your customer base annually. You are simply over-taxing your remaining happy customers to cover the holes left by the angry ones leaving.
Investors and acquirers have grown wise to this. In 2025, while NRR drives the upside of your valuation, Gross Revenue Retention (GRR) sets the floor. If your GRR is below 85%, you don't have a growth business; you have a leaky bucket that requires exponentially more fuel (CAC) to keep full. You cannot upsell your way out of a churn problem.
Scaling Sarah often incentivizes her CS team on NRR. This seems logical: grow the account. But without a counter-balancing metric on GRR (logo retention and revenue retention sans expansion), the team will naturally gravitate toward the "friendly" accounts—the ones easy to upsell—while neglecting the quiet accounts that are slowly drifting toward cancellation.
When you finally go to market or seek Series C funding, the diligence team will strip out the upsells. They will look at the cohorts. And if they see a GRR of 82%, they will know that your product is not sticky, your onboarding is failing, or your competitive moat is dry.
Let's strip away the anecdotal evidence and look at the data. The market has bifurcated in 2025. The "growth at all costs" era is dead; the "efficient growth" era is here, and retention is its currency.
According to Software Equity Group's 2024/2025 research, companies with NRR above 120% are trading at a 63% valuation premium over the market median. That is the difference between a 4x exit and a 7x exit.
However, that premium only holds if the foundation is solid. Based on ChartMogul's 2025 SaaS Growth Report and cross-referenced data from ICONIQ Growth, here are the benchmarks you must measure against:
If your GRR is 90% and your NRR is 105%, you have a stable business with weak expansion motions. You need better pricing packaging or add-on modules. This is a fixable "offense" problem.
If your GRR is 75% and your NRR is 105%, you are in a turnaround scenario. You are burning through your Total Addressable Market (TAM). Every customer who churns is a detractor who will never buy from you again. Expansion revenue from the remaining 75% is a temporary drug masking a terminal illness.
This dynamic explains why median SaaS firms often trade at a discount despite decent growth. As detailed in The New Rule of 40, investors are prioritizing durability. A dollar of expansion revenue is high-margin (low CAC), but a dollar of churned revenue is destructive because replacing it costs 5x-25x more than keeping it.
For enterprise B2B firms (ACV >$50k), the expectations are even higher. ICONIQ Growth's State of Software 2025 indicates that top-performing enterprise companies maintain GRR rates above 95%. If you are selling mission-critical software and losing 10% of your revenue base annually (excluding upsells), the market assumes your product is not actually mission-critical.
If you suspect you are masking a churn problem with upsells, stop looking at the blended average. You need to perform a cohort diagnostic immediately. Here is the operator's playbook for correcting the imbalance.
Stop reporting just "Retention" or "NRR." Present them side-by-side. Use this narrative: "GRR is our defense; NRR is our offense." If defense is failing, pause aggressive expansion targets and fix the hole in the boat. Read more on how to frame this in NRR Below 100? Your Customer Success Function Is Broken.
Your Customer Success Managers (CSMs) cannot effectively hunt and farm at the same time if the farm is on fire. If GRR is below 90%:
Often, low GRR manifests as "down-sells"—customers reducing seat counts or removing modules—rather than full cancellations. This is the silent killer. It indicates your software is shelf-ware. Implement a "Usage-to-Contract" audit. If a customer is utilizing 40% of what they bought, they will churn or down-sell at renewal. Preempt this by rightsizing them early in exchange for a longer term, or launching a targeted adoption campaign 6 months before renewal.
Low GRR frequently stems from sales closing the wrong customers. Look at the churned logos from the last 12 months. Is there a pattern? (e.g., "We lose every customer under $10M revenue" or "We lose every customer who uses the legacy integration"). If so, fire that segment. Stop selling to them. It is better to have slower growth with 95% GRR than fast growth with 80% GRR. See Stop Bleeding Revenue for industry-specific churn thresholds.
NRR is how you get rich; GRR is how you survive long enough to get there. In the current market, valuation multiples accrue to the survivors. secure the hull first, then turn on the engine.
