What Is a Good NRR for SaaS? (2026 Benchmarks by Stage)

TLDR: Net revenue retention (NRR) is the single best one-number summary of SaaS business health. The benchmarks in 2026:

• Best-in-class public SaaS: 130%+
• Public SaaS median (at IPO): ~120%
• Private SaaS median: ~105%
• Under 100% NRR: structurally bleeding

But the more important point: NRR can lie. Optimize gross retention (GRR) first. NRR is the lagging summary. GRR is what's actually under your control.

I've watched 3 SaaS teams hit 110% NRR while their GRR was under 85%. Two of them are dead. One pivoted hard. NRR can hide a leaky bucket for years until expansion stops compounding.

What is net revenue retention (NRR)?

NRR is the percentage of recurring revenue you retain from existing customers over a period, including expansion (upgrades, added seats, usage growth) and subtracting downgrades and churn.

The formula:

NRR = (starting MRR + expansion MRR − downgrade MRR − churned MRR) / starting MRR

Critical: NRR measures existing customers only. New logos do not count. If you mix in new ARR, you're computing growth rate, not net retention.

NRR above 100% means your existing base is growing on its own. Below 100% means it's shrinking and your sales team has to outrun the leak just to stay flat.

What is a good NRR for SaaS?

The 2026 benchmarks, by stage:

The category leaders in 2026 (Snowflake, Datadog, Toast, and a handful of others) routinely report NRR above 130%. The bar for what counts as "great" has climbed about 5-10 points in the last 3 years as buyers tightened.

For context: Bessemer's State of the Cloud and the OpenView annual reports are the cleanest public benchmark sources.

What is the difference between NRR and GRR?

GRR (gross revenue retention) is what you held onto. GRR cannot exceed 100%. The formula:

GRR = (starting MRR − downgrade MRR − churned MRR) / starting MRR

NRR is GRR plus expansion. Expansion can push NRR over 100%. GRR by definition cannot.

Why this matters: GRR shows the structural quality of the customer base. NRR shows GRR after the expansion sales team has done its work. If GRR is 92% and NRR is 115%, you have a healthy core that's growing. If GRR is 80% and NRR is 110%, you have a leaky bucket that's currently masked by expansion. That's a worse business than the first one, even though the headline NRR is slightly higher.

Most investors and operators I talk to in 2026 look at both. The pattern that wins: GRR above 90%, NRR above 115%, both improving year-over-year.

Should you optimize NRR or GRR first?

GRR. Always GRR first.

Three reasons:

1. Expansion has a ceiling. Customers can only buy so many seats, upgrade so many tiers, expand so much usage. Eventually expansion slows. If your GRR is bad, you'll find that out when expansion slows because there's nothing underneath holding the business up.
2. Churn compounds badly. A 15% gross churn rate means you lose 80% of a cohort over 10 years. That math gets uglier as you scale.
3. Fixing GRR is cheaper than fixing NRR. Most GRR fixes (dunning, save flows, behavioral onboarding) are tools or experiments under $1000/month and pay back in 30 days. Driving NRR usually means redesigning packaging or building usage-based pricing, which is months of work.

The order of operations I recommend to every SaaS team:

1. Fix involuntary churn first. See what is involuntary churn for the playbook. Drops 15-25% of total churn in 30 days.
2. Ship a structured save flow. Recovers 10-20% of voluntary cancellations.
3. Build a customer health score and route to CS. See customer health score guide.
4. Only then start optimizing expansion plays for NRR growth.

How do you calculate NRR correctly?

Standard formula:

NRR = (starting MRR + expansion MRR − downgrade MRR − churned MRR) / starting MRR

Common errors that wreck the number:

• Including new logos. New customers do not count in NRR. They're growth, not retention.
• Using bookings instead of revenue. Bookings include deals not yet recognized. Use recognized MRR or ARR.
• Mixing annual and monthly without normalizing. If you have both contract types, convert to MRR-equivalent or your number won't be comparable across periods.
• Computing on too short a window. NRR should be measured on at least a 12-month cohort to smooth out timing noise. Monthly NRR is mostly noise.

The cleanest version is dollar-based net retention: pick a cohort of customers active at the start of the period, sum their ARR at month 0, sum their ARR at month 12, divide. That's the truest number.

For tools to compute this automatically, ProfitWell (now Paddle), ChartMogul, and Baremetrics are the standard picks. See our ProfitWell vs Baremetrics comparison if you're picking between them.

What drives NRR up?

Two levers, and they move at different speeds:

Expansion (faster lever)

• Usage-based pricing: customers pay more as they use more, automatically
• Tiered packaging that nudges accounts up tiers as they grow
• Seat-based expansion plays: making it easy to add seats inside the product
• Cross-sell of complementary modules to existing customers

Expansion plays usually move NRR within 1-2 quarters of shipping.

Contraction reduction (slower but more durable lever)

• Lower churn (the GRR work above)
• Fewer downgrades through better activation and usage-based engagement
• Better save flows that retain at-risk accounts at lower tiers instead of losing them entirely

Contraction work moves NRR slower but the gains compound over years instead of quarters.

What is the benchmark NRR for B2B vs B2C SaaS?

B2C SaaS structurally has lower NRR because the expansion ceiling per customer is lower. A typical B2C subscriber can't really expand. They either keep paying or they leave.

• B2B SaaS: 105-130% NRR is the normal range. Best-in-class above 130%.
• B2C SaaS: 80-100% NRR is the normal range. Best-in-class above 100%.
• Subscription ecommerce: 70-90% NRR is normal. Hard ceiling because customers don't add seats.
• Enterprise SaaS (5+ figure ACV): 110-140% NRR. Highest range because expansion is structural.

Comparing B2B SaaS NRR to subscription box NRR is apples-to-oranges. Benchmark within your segment.

Where should you start?

If you don't know your NRR today, that's the first thing to fix. The benchmarks above are useless without knowing where you stand.

To find your number fast:

• Use the MRR Impact Simulator to see how each 1% of NRR change translates to dollars at your scale.
• Take the Churn Health Check to find your weakest retention spots.
• See churn rate benchmarks for more cuts of the data by industry and company size.

If your NRR is under 100%, fix GRR first. If your NRR is 100-110%, you're roughly normal for stage. If you're above 120%, you're in the top quartile and the question becomes how to defend that position as you scale.