---
title: "Net Revenue Retention Benchmarks for SaaS: What 'Good' Actually Looks Like in 2026"
description: "Discover what good NRR actually looks like in 2026. Enterprise, mid-market, and SMB benchmarks that actually match your ACV tier."
author: "James"
author_title: "Head of Content"
author_url: "https://blog.startuplist.tech/author/james"
date: 2026-10-14T12:06:37.283Z
language: "en"
category: "SaaS Growth"
tags: ["SaaS net revenue retention benchmarks", "NRR by ACV tier", "enterprise SaaS NRR 2026", "mid-market SaaS benchmarks", "net revenue retention targets"]
canonical: "https://blog.startuplist.tech/net-revenue-retention-benchmarks-for-saas-what-good-actually"
---

# Net Revenue Retention Benchmarks for SaaS: What 'Good' Actually Looks Like in 2026

Good SaaS net revenue retention in 2026 depends entirely on your customer segment — enterprise SaaS (>$100K ACV) should be landing around 115-118%, mid-market ($25K-$100K ACV) around 104-108%, and SMB somewhere between 90% and 100%. If someone tells you "good NRR" without asking your ACV tier first, they're guessing.

That's the honest answer buried under a decade of conflicting LinkedIn posts. Some say 100% is fine. Others say anything under 130% is a red flag. Both are technically right, depending on who's in the sample. This post reconciles the mess and gives you a number to actually anchor to.

## Why every "good NRR" number you've seen contradicts the last one

Here's the trick nobody tells you: every benchmark report blends a different mix of company stages, ACV tiers, and pricing models, then publishes one median as if it's universal truth.

Optifai's 2025-2026 dataset (939 companies) puts the overall median at 102%, but splits enterprise at 115-125% and SMB at 90-105% — a spread wide enough to make the "one number" approach useless. Digital Applied found a similar pattern: a 21-point gap between enterprise (118%) and SMB (97%) in their 2024 data. Meanwhile Optifai's earlier ChartMogul-sourced figure (N=2,100 venture-backed companies) puts the *general* median at 106%, describing >130% as best-in-class and sub-100% as concerning.

None of these are wrong. They're measuring different populations.

> 💡 Anchor to three numbers before you anchor to a benchmark: your ACV tier, your pricing model (seat vs usage-based), and your GRR. Call it the Three-Number Rule. Skip any one of them and the "good NRR" figure you're comparing yourself to is meaningless.

If you're a $40K-ACV mid-market tool, comparing yourself to a $250K-ACV enterprise vendor's 118% NRR will make you feel like you're failing when you're actually on target. The reverse is just as damaging — enterprise founders benchmarking against blended medians end up setting targets too low for their comp set.

## The segment-by-segment numbers to actually use

Strip out the noise and the 2025-2026 data converges more than it first appears. Here's the clearest read across sources:

| Segment | ACV range | Typical NRR | Source |
|---|---|---|---|
| Enterprise | >$100K | 115-118% | Digital Applied, SaaS Mag |
| Mid-market | $25K-$100K | 104-108% | Ryan Allis (2026), SaaS Mag |
| SMB | <$25K | 90-100%, sometimes below | Optifai, Directive Consulting |

Ryan Allis's 2026 SaaS Benchmarks Report (LinkedIn) lines up closely: SMB at 100%, mid-market at 104%, enterprise at 115%. It also flags why SMB lags — annual account churn runs around 15% for SMB vs roughly 10% for mid-market. That's not a pricing problem. It's a base-rate problem: smaller customers have shorter attention spans, smaller budgets, and higher turnover in the buying seat itself.

The market-wide median has also moved. Digital Applied's analysis shows median private SaaS NRR falling from roughly 105% in 2021 to about 101% in 2024. If you're benchmarking against a 2021-era comp set — which a lot of founders raising now unconsciously are, because that's when their investors last updated their mental model — you're holding yourself to a bar that no longer reflects the market. 101-102% is the market median today. 100% is the healthy floor. Anything above 110% for mid-market, or above 115% for enterprise, is genuinely top-tier — not just "fine."

## NRR alone lies to you — pair it with GRR

Here's the piece almost every benchmark table skips: NRR without gross revenue retention (GRR) is half a story.

A company posting 112% NRR sounds great until you learn their GRR is 82%. That means they're losing 18% of revenue to [churn and downgrades](/b/startuplist-blog/why-saas-churn-spikes-in-the-first-90-days-after-signup-and) every year, then papering over it with aggressive upsells to whoever's left. It's a leaky bucket getting refilled fast enough that the headline number looks healthy. Compare that to a company at 105% NRR with 95% GRR — smaller expansion, but a far more stable base.

VCs read both numbers together for exactly this reason. High NRR + high GRR is the gold standard: you're not losing customers, and the ones who stay spend more. High NRR + low GRR is a warning sign dressed up as good news. High GRR + flat NRR (100%) means you've solved churn but haven't cracked expansion — a real problem, but a much more fixable one than mass churn.

> ⚠️ If your NRR is above 110% and you can't immediately state your GRR, find out before your next board meeting. Investors will ask, and "I don't track that separately" reads as a red flag, not a shrug.

## Usage-based pricing is the quiet NRR cheat code

The single biggest lever separating today's NRR winners from the pack isn't sales motion or CS headcount — it's [pricing model](/b/startuplist-blog/the-complete-guide-to-saas-pricing-strategy-for-2026-models).

The Aleph x Benchmarkit 2026 benchmark found usage-based companies averaging 108% NRR against 98% for seat-based peers. FE International's analysis with m3ter found an even wider gap: 115-130% for usage-based models vs 95-105% for flat-rate subscriptions. The mechanism is simple — usage-based revenue expands automatically as customers grow their consumption, with zero sales motion required. A customer processing 2x the transactions this quarter pays 2x, without anyone picking up a phone.

Seat-based pricing caps your expansion at headcount growth, and headcount growth is slow, political, and budget-constrained. Usage-based pricing ties your revenue directly to the thing your product actually does. That's not a minor edge — it's a structurally different growth curve.

This matters more heading into 2026 than the benchmark reports currently model.

## The AI-agent risk nobody's pricing into these benchmarks

Every NRR benchmark published so far assumes seat counts grow, or at worst stay flat. That assumption is getting shakier by the quarter.

If AI agents start doing the work that used to require five seats of a workflow tool, seat-based SaaS products face a structural headwind that has nothing to do with customer satisfaction or product quality. A customer can love your tool, renew happily, and still cut their seat count in half because an agent now handles the junior-analyst tasks three people used to do manually. That shows up as negative expansion — or outright contraction — inside accounts that would previously have counted as healthy renewals.

This is a live risk for SMB and mid-market tools specifically, where seat counts are smaller and more sensitive to headcount changes. Enterprise buyers tend to move slower on this kind of restructuring, which may widen the enterprise-vs-SMB NRR gap even further than the current 20+ point spread.

The practical takeaway: if your pricing is seat-based and your roadmap includes AI features that make individual users more productive, you're quietly building a product that shrinks your own NRR. Usage-based or outcome-based pricing sidesteps this entirely — more AI-driven output usually means more usage, not less.

## How to actually move your NRR up a tier

Most benchmark content stops at the table. Here's what actually moves the number.

- **Fix GRR first.** Expansion on a leaky base is wasted effort. If GRR is below 90%, your CS motion — not your upsell motion — is the priority.
- **Segment your renewal motion by ACV.** SMB needs automated, low-touch nudges (usage alerts, in-app upgrade prompts). Enterprise needs a named CSM running a structured QBR cadence. Applying the SMB playbook to enterprise accounts (or vice versa) wastes budget.
- **Price for expansion, not just acquisition.** If you're seat-based, add a usage or consumption layer even if it starts small. A hybrid model captures upside that flat seats never will.
- **Track cohort NRR monthly, not annually.** Annual NRR hides the quarter where things started slipping. By the time the annual number moves, you're three quarters behind the fix.
- **Skip the metric if you're pre-$1M ARR.** With 10-20 customers, one enterprise logo churning swings your NRR by double digits. Track logo retention and qualitative expansion signals instead — NRR becomes statistically meaningful somewhere past 50-100 customers.

One more terminology note worth flagging in your next board deck: NRR and Net Dollar Retention (NDR) are the same metric under different names. Investors use them interchangeably. Don't let a founder get caught explaining a "discrepancy" that's just a labeling difference.

**Does UK SaaS NRR differ from US benchmarks?**

No large-scale UK-specific NRR benchmark study currently exists — nearly all published data (ChartMogul, Benchmarkit, SaaS Mag) is US-weighted. In practice, UK B2B SaaS founders should use the same ACV-tier framework above; pricing model and segment matter far more than geography for this particular metric.

## Why this number moves your valuation, not just your board deck

NRR isn't just a retention metric anymore — it's a direct valuation input. McKinsey's analysis of 55 B2B SaaS companies, cited by Levera Partners, found top-quartile-valued companies posting around 113% NRR against 98% for bottom-quartile peers. That 15-point gap corresponded to roughly a fivefold difference in valuation multiple. That's not a rounding error — that's the difference between a mediocre exit and a great one.

McKinsey's broader analysis also found that high-NRR SaaS businesses held their valuations better through market downturns than high-growth, leaky-retention peers. In a market where growth-at-all-costs has fallen out of favor with investors, NRR has become the metric that separates durable businesses from momentum stories. If you're heading into a raise or an exit conversation in the next 12-18 months, your NRR quartile is doing more work in that data room than your growth rate slide.

Know your segment, pair NRR with GRR before you quote either number, and set your target from your own ACV tier and pricing model — not from whichever benchmark post crossed your feed this week.

## Frequently Asked Questions

### What's the difference between NRR and NDR?

Nothing — they're the same metric with different names. Net Revenue Retention and Net Dollar Retention are used interchangeably by investors and founders, so don't be thrown off if a term sheet or board deck swaps between them.

### What NRR is too small to measure reliably?

Below roughly 50-100 customers, cohort sizes are too small for NRR to be statistically stable — one enterprise churn event can swing the number by double digits. Pre-$1M ARR companies should track logo retention and qualitative expansion signals instead, and revisit formal NRR tracking once the customer base scales.

### Is 100% NRR good or bad?

100% NRR means your existing customer base neither grew nor shrank in revenue — it's the healthy minimum, not a target to celebrate. For SMB-focused companies it's roughly in line with the market median; for mid-market or enterprise businesses, 100% signals you're leaving expansion revenue on the table.

### How does pricing model affect NRR benchmarks?

Usage-based pricing models post materially higher NRR — roughly 108-130% versus 95-108% for seat or flat-rate models — because revenue scales automatically with customer usage rather than requiring a sales-led upsell motion. If you're transitioning pricing models mid-lifecycle, expect a temporary NRR dip during the switch before the usage-based expansion kicks in.

### Why did median SaaS NRR fall since 2021?

Digital Applied's data shows the median dropping from roughly 105% in 2021 to about 101% in 2024, largely reflecting tighter budgets, slower seat growth, and more scrutinized renewals across the broader SaaS market post-2022. Founders benchmarking against older comp sets should recalibrate — 101-102% is the current market median, not a warning sign.

**Check out the latest startups**

[Visit StartupList](https://startuplist.tech)

## Written by

**[James](https://blog.startuplist.tech/author/james)**, Head of Content