Net revenue retention benchmarks: the formula, 2026 data, and how to push past 100%
Net revenue retention (NRR) measures how much recurring revenue a fixed group of existing customers generates a year later, after expansion, downgrades, and cancellations, counting zero new logos. In 2026 the all-B2B median sits near 106 percent. That number splits hard by segment: SMB-focused SaaS lands around 97 percent, mid-market around 108 percent, enterprise around 118 percent, and the best clear 130 percent.
NRR is the cleanest single read on whether your existing base is a growth engine or a leaky bucket.
It is also the first metric most investors check. A 10-point gain can add 20 to 30 percent to a company’s valuation with no change to ARR or growth rate.
What net revenue retention actually measures
NRR answers one question: if you stopped signing new customers tomorrow, would your revenue grow or shrink? You take a cohort of customers, freeze it on day one, and measure what that exact group is worth twelve months later.
The trick is in what you exclude. You never count revenue from customers acquired during the period.
Three forces move the number:
- Expansion: upsells, cross-sells, added seats, usage growth.
- Contraction: downgrades and partial reductions, also called revenue churn on the soft side.
- Churn: customers who cancel outright.
Above 100 percent means expansion outweighs everything you lost. Below 100 percent means your base is quietly shrinking, and you are paying for new acquisition just to stand still.
The NRR formula, with a worked example
The NRR formula is straightforward once you fix the cohort:
NRR = (Starting MRR + Expansion MRR − Contraction MRR − Churned MRR) / Starting MRR × 100
Measure the same customers at the start and end of the period. Never add revenue from new customers signed during it. If you need a refresher on the base unit, here is how MRR is defined.
Work an example. A cohort of existing customers starts the year at $500,000 in MRR:
- Expansion from upgrades and seat growth: $90,000
- Contraction from downgrades: $20,000
- Churned MRR from cancellations: $40,000
Run it:
($500,000 + $90,000 − $20,000 − $40,000) / $500,000 × 100 = $530,000 / $500,000 × 100 = 106 percent
That cohort is worth 6 percent more than it was a year ago, with no new customers. If you want to skip the spreadsheet, the net revenue retention calculator runs the same math.
One note on time frames. Most teams report a trailing-12-month cohort to smooth seasonality. A monthly NRR of 100.8 percent compounds to roughly 110 percent over a year, so always state whether you mean monthly or annual.
NRR vs gross revenue retention: the gap is the signal
Gross revenue retention (GRR) uses the same starting cohort but excludes expansion entirely. It only subtracts losses:
GRR = (Starting MRR − Contraction MRR − Churned MRR) / Starting MRR × 100
Because GRR ignores upsells, it can never exceed 100 percent. It is the floor.
Using the example above, GRR is ($500,000 − $20,000 − $40,000) / $500,000 = 88 percent. So that cohort leaks 12 points a year before any expansion refills the bucket.
The gap between the two metrics is where the real story lives.
| Metric | What it counts | Ceiling | What it tells you |
|---|---|---|---|
| GRR | Losses only (churn + contraction) | 100 percent | How leaky the bucket is |
| NRR | Losses plus expansion | None | Whether expansion refills it |
A company with NRR at 110 percent and GRR at 90 percent looks healthy on the headline number. But the 20-point spread means strong upsells are masking real logo churn.
Investors read that gap as product stickiness versus sales-led expansion. A tight gap (GRR 95, NRR 108) is a far better business than a wide one (GRR 85, NRR 108), even at identical NRR.
2026 NRR benchmarks by segment
There is no single good NRR. There is only a good NRR for your contract size, and the difference is large enough to misprice a company. Retention rises with deal value because bigger contracts are stickier and have more room to expand.
Here is where the 2026 data lands, drawn from the SaaS Capital 2025 retention survey and public filings.
| Segment | Typical ACV | Median NRR | Read |
|---|---|---|---|
| SMB-focused | Below $25K | approx 97% | At benchmark, but base is shrinking |
| Mid-market | $25K to $100K | approx 108% | Healthy expansion |
| Enterprise | Above $100K | approx 118% | Strong, sticky base |
| Best-in-class | Any | above 130% | Durable negative churn |
| All-B2B median | Mixed | approx 106% | The blended midpoint |
The $25K to $50K band shows the spread clearly: median 102 percent, top quartile 111 percent, bottom quartile 97 percent. Bootstrapped SaaS in the $3M to $20M ARR range ran a median NRR of 103 percent against GRR of 91 percent in SaaS Capital’s 2026 data.
Most posts get this part wrong. A 97 percent NRR is on-target for an SMB product and a crisis for an enterprise one. Compare inside your ACV band, never against a single global number.
What investors expect by ARR stage
At Series A, investors treat 100 percent NRR as the baseline, 110 to 120 percent as competitive, and anything above 120 percent as premium. Top-quartile Series A companies cluster around 99 to 110 percent. By growth stage, the top quartile crosses 105 percent and the strongest hit 110 to 120 percent.
Bessemer’s shorthand is the framework most growth investors carry in their heads:
- 100 percent NRR is good
- 110 percent is better
- 120 percent and above is best
That scale was built for growth-stage enterprise SaaS, not SMB, which is why segment context matters before you apply it.
KeyBanc and Sapphire Ventures’ 2025 survey of roughly 100 private SaaS firms put median net retention near 101 percent, with top performers around 104 to 106 percent. Snowflake, a public best-in-class marker, reported NRR near 125 to 126 percent through FY2025.
NRR also predicts growth. SaaS Capital found companies at or above 110 percent NRR grow faster than the 24 percent population median, while those below 100 percent grow slower. Retention is a growth multiplier, not just a defensive metric.
Pair it with the SaaS quick ratio to see whether your overall engine, including new logos, is efficient.
Negative churn: how NRR climbs above 100 percent
Negative churn is the condition where expansion revenue from existing customers exceeds everything you lost to downgrades and cancellations. When that happens, NRR breaks above 100 percent and your installed base grows on its own, even if you never sign another account.
This is the structural prize in SaaS. Revenue compounds without new acquisition spend, so growth gets cheaper every year. A base running 120 percent NRR doubles roughly every four years before a single new customer is added.
You get there by widening the gap between expansion and loss. The levers that move it:
- Usage- or seat-based pricing so revenue grows as customers grow, without a sales motion.
- Pricing tiers with clear upgrade paths, so expansion is a product event, not a negotiation.
- Better onboarding that gets customers to value fast, which cuts early churn and sets up expansion.
- Proactive saves on low-engagement accounts before they downgrade.
Expansion revenue is the only NRR input you can grow without limit, so most of the work lives there.
The valuation math: why 10 points is worth so much
A 10-point NRR improvement, say 110 to 120 percent, can translate to a 20 to 30 percent valuation uplift at identical ARR and growth. The reason is compounding. Higher expansion lowers your reliance on new-logo acquisition, which makes future revenue more predictable and cheaper to produce.
That makes NRR less a retention stat and more the single highest-leverage valuation input a founder controls.
You can spend two quarters lifting acquisition and move the multiple a little. Lift NRR ten points and the same ARR is worth meaningfully more, because the market is pricing the durability of the base, not just its size.
This is also why the metric gets gamed, intentionally or not.
How teams accidentally inflate NRR
The most common error is letting new customers leak into the cohort. If anyone acquired during the period shows up in your ending revenue, your NRR is fiction. Freeze the cohort on day one and measure only those accounts.
Other ways the number quietly flatters you:
- Blending segments. A 106 percent blended figure can hide an SMB segment bleeding at 95 percent, propped up by enterprise at 120 percent. The blend looks fine while half the base is on fire.
- Mixing annual prepay into MRR. Count normalized recurring revenue, not a lump-sum annual invoice booked in one month.
- Reporting monthly NRR as if it were annual. A 100.8 percent monthly figure is roughly 110 percent annualized. Quoting the monthly number understates a strong base, and the annual number flatters a weak one.
Watch the GRR gap too. NRR above 100 percent with gross retention near 88 percent means expansion is masking serious logo churn. The headline reads healthy. The base does not. Tracking churn rate alongside NRR keeps you honest about which is which.
Tracking segmented NRR in real time
The reason NRR is hard to get right is the decomposition. You need clean expansion, contraction, and churn split out of raw billing data, by cohort, with new customers excluded. Doing that by hand in a spreadsheet is where most of the errors above creep in.
Mowt computes NRR directly from raw Stripe data in real time, segmented by ARR or ACV cohort, so you see which slice of your base is dragging the median instead of staring at one blended number. That turns NRR from a quarterly board-deck figure into something you can act on while the quarter is still open. It pairs with cohort analysis to show whether retention is improving for newer cohorts or decaying.
Pull your last twelve months, freeze the cohort, and split it by segment. The number that matters is not your blended NRR. It is the worst segment you have, and whether it is trending up.
FAQ
What is a good net revenue retention rate in 2026?
It depends entirely on your segment. The all-B2B median is around 106 percent, but a good benchmark for SMB-focused SaaS is roughly 97 to 100 percent, mid-market around 108 percent, and enterprise around 118 percent. Above 120 percent is excellent and above 130 percent is best-in-class, so always compare within your ACV band rather than against one global number.
What is the difference between NRR and GRR?
Gross revenue retention counts only the revenue you keep, subtracting churn and downgrades, so it can never exceed 100 percent. Net revenue retention starts from the same base but adds expansion, so it can rise above 100 percent. The gap between them shows whether upsells are masking real logo churn.
How do you calculate net revenue retention?
Take the starting MRR of a fixed cohort of existing customers, add expansion, subtract contraction and churn, divide by the starting MRR, and multiply by 100. Only measure customers who existed at the start of the period, and never count revenue from new customers signed during it. Most teams use a trailing-12-month cohort to smooth seasonality.
Can NRR be above 100 percent, and what is negative churn?
Yes. NRR exceeds 100 percent whenever expansion from existing customers outweighs revenue lost to downgrades and cancellations, a condition called negative churn. Your installed base then grows in revenue with no new customers, which is why best-in-class companies run NRR of 120 to 130 percent or higher.
What NRR do investors expect at Series A or growth stage?
At Series A, investors treat 100 percent NRR as the baseline, 110 to 120 percent as competitive, and anything above 120 percent as premium, with top-quartile companies around 99 to 110 percent. By growth stage the top quartile crosses 105 percent and the strongest hit 110 to 120 percent.
About the Author
Matt Smith
Serial entrepreneur and former big 4 consultant turned SaaS operator. Built and scaled analytics and data warehouses platforms at multiple enterprise Stripe companies before founding Mowt. Passionate about making complex metrics accessible to every founder.