Net Revenue Retention (NRR): How to Calculate and Improve It

Net revenue retention (NRR) measures how much recurring revenue you keep and grow from existing customers over a given period, after accounting for expansions, contractions, and churn. Calculate it by dividing end-of-period MRR from your existing cohort by their starting MRR. A strong SaaS company targets NRR above 110%, which means the installed customer base generates more revenue each period without any new sales.

This metric has taken on new urgency in 2026. SaaStr's recent analysis of renewal discounting highlights a tension every SaaS operator faces: saving a churning customer with a steep discount inflates your logo retention but can quietly erode the revenue quality that NRR is designed to measure. Understanding NRR, and the components that move it, helps you make that tradeoff with open eyes.

What Is Net Revenue Retention?

Net revenue retention is the percentage of recurring revenue retained from existing customers after factoring in expansion, contraction, and churn. It answers a specific question: if you stopped acquiring new customers today, would your revenue grow, hold flat, or shrink?

NRR differs from gross revenue retention (GRR) in one critical way. GRR only captures losses, so it caps at 100%. NRR includes expansion revenue (upsells, cross-sells, seat additions, usage growth), which means it can exceed 100%. That distinction matters. A company with 85% GRR and 115% NRR has a churn problem that expansion revenue is currently masking. Both numbers are useful, but NRR tells you the net trajectory.

How to Calculate Net Revenue Retention

The standard formula:

NRR = ((Starting MRR + Expansion MRR - Contraction MRR - Churned MRR) / Starting MRR) x 100

Here's each component:

• Starting MRR: the total monthly recurring revenue at the beginning of the period from the cohort you're measuring
• Expansion MRR: additional revenue from existing customers (upgrades, seat additions, usage overages)
• Contraction MRR: revenue lost from existing customers who downgraded but did not cancel
• Churned MRR: revenue lost from customers who canceled entirely

Worked example: Say you start the quarter with $200K MRR from 150 customers. Over the quarter, those same 150 customers generate $18K in expansion MRR, $4K in contraction, and $7K in churn. Your NRR is:

NRR = (($200K + $18K - $4K - $7K) / $200K) x 100 = 103.5%

That 103.5% means your existing base is growing revenue at 3.5% per quarter on its own, roughly 14.5% annualized, before a single new customer walks in the door.

One important caveat: NRR should be calculated on a cohort basis, using the same set of customers at the start and end of the period. If you mix in new customers acquired during the period, the number loses its diagnostic value.

Calculate Your NRR

NRR Benchmarks by Stage and Segment

NRR expectations vary significantly by who you sell to. Smaller customers churn at higher rates and expand less, so SMB-heavy companies structurally run lower NRR than enterprise-focused ones.

Segment	Below Average	Average	Strong	Elite
SMB (< $1K ACV)	< 85%	85-95%	95-105%	> 105%
Mid-Market ($5K-$50K ACV)	< 95%	95-105%	105-115%	> 115%
Enterprise ($50K+ ACV)	< 100%	100-110%	110-130%	> 130%

A few patterns worth calling out. First, NRR tends to improve as companies mature because the customer base shifts toward larger, stickier accounts. A Series A company with 95% NRR is not necessarily in trouble. It may simply reflect an early product with limited upsell surface area.

Second, usage-based pricing models often produce higher NRR than seat-based models because customers expand organically without a procurement decision. If you're building a consumption-based revenue forecast, factor this NRR advantage into your projections.

Why NRR Matters More Than Acquisition Alone

Here's the math that makes NRR so powerful. A company with $5M ARR and 120% NRR will generate $6M from existing customers next year, all without spending a dollar on acquisition. A company with the same $5M ARR but 90% NRR will retain only $4.5M, meaning it needs $500K in net new ARR just to stay flat and far more to grow.

This is why investors treat NRR as one of the strongest signals of durable growth. High NRR compounds. At 120% annual NRR, a $5M cohort becomes $12.4M in five years from expansion alone. At 90%, that same cohort shrinks to $3.3M. The gap widens every year.

For financial modeling, NRR is the multiplier that either amplifies or undermines your acquisition investments. If you're building a SaaS financial model, NRR should be a first-class input, not something buried inside a blended churn assumption. Model expansion and contraction as separate line items so you can stress-test scenarios: what happens to revenue if expansion slows 20%? What if you cut discounting and lose 5% more logos but keep price integrity?

The Discount Trap: Retention at What Cost?

SaaStr recently made the case that discounting to save a churning renewal is often worth it, even when the math says otherwise. The argument: a retained customer, even at a lower price, generates referrals, keeps your logo count healthy, and leaves the door open for future expansion. The piece calls this "the NRR trap" because purists would say the discount erodes revenue quality.

Both sides have merit, and the honest answer depends on context. Here's the tradeoff in practice:

• Discount to save: you retain the logo and a reduced revenue stream. Your churn rate improves, but your NRR absorbs the contraction hit.
• Let the customer churn: your churn rate takes the hit, but your remaining revenue base stays at full price. Your NRR takes a different kind of hit (via churned MRR), but the average revenue quality of your base stays intact.

The question to ask is whether the customer is likely to expand back to full price within 6-12 months. If yes, the discount is a bridge. If no, you're permanently compressing ARPU on that account, which drags down NRR for every future period.

For founders building financial models, the practical implication is this: don't model NRR as a single number. Segment it by customer tier and by whether accounts are on discounted vs. standard pricing. That granularity reveals whether your headline NRR is healthy expansion or discount-inflated retention.

How to Improve Net Revenue Retention

Build Expansion Into the Product

The highest-NRR companies don't rely on sales teams to drive upsells. They build expansion triggers into the product: usage limits that encourage plan upgrades, feature tiers that unlock as teams grow, and collaboration tools that pull in more seats organically. If customers naturally use more of your product over time, NRR rises without a single outbound email.

Reduce Contraction Before Chasing Expansion

Contraction is the silent NRR killer. A customer who downgrades from $500/month to $200/month has effectively "churned" 60% of their revenue. Focus on understanding why customers downgrade. Common reasons include underutilization of features (an onboarding problem), seasonal business fluctuations (a billing flexibility problem), and perceived lack of ROI (a value delivery problem).

Segment Your Retention Metrics

Aggregate NRR hides patterns. Break your churn and retention data into segments by customer size, vertical, acquisition channel, and tenure. You may discover that enterprise accounts have 125% NRR while SMB accounts sit at 80%. That's not one problem. Those are two completely different problems with different solutions.

Track Leading Indicators

NRR is a lagging metric. By the time you see it drop, the damage is months old. Track leading indicators instead: product engagement scores, support ticket trends, login frequency, and feature adoption rates. These signals give you a chance to intervene before contraction or churn events materialize. Pair these with your burn rate monitoring so you can spot revenue erosion before it affects your runway projections.

Key Takeaways

• NRR above 100% means your existing customer base grows revenue on its own, without any new acquisition. Below 100% means you're filling a leaky bucket.
• Calculate NRR on a cohort basis using starting MRR, expansion, contraction, and churn for the same set of customers over a defined period.
• Enterprise NRR benchmarks run 110-130% while SMB targets are 95-105%. Segment matters more than stage.
• Discounting to save renewals is a valid tactic, but model it separately so you can see whether your NRR is built on genuine expansion or compressed pricing.
• Build NRR as a first-class input in your SaaS financial model, not a derived number buried in blended assumptions.

Net revenue retention tells you whether your product is a growth engine or a leaky bucket. Model your NRR alongside your MRR waterfall in Revenue Map to see exactly where expansion and contraction are pulling your trajectory, and stress-test what happens when those inputs change.