Net revenue retention (NRR) vs GRR in SaaS valuation: expansion growth versus the retention floor

The two formulas answer different questions

Stripe's current retention guide defines the distinction clearly. NRR includes the negative effects of churn and contraction plus the positive effects of expansion. GRR excludes expansion and measures only what the starting base retained.

In plain language:

• GRR: Of the recurring revenue we began with, how much survived without counting extra sales?
• NRR: After losses and expansion inside that same customer cohort, how much recurring revenue do we have now?

A simplified calculation is:

GRR = (starting recurring revenue − churn − contraction) ÷ starting recurring revenue

NRR = (starting recurring revenue − churn − contraction + expansion) ÷ starting recurring revenue

To calculate net revenue retention, apply the second formula to one customer cohort. To calculate GRR from the same cohort, drop the expansion term.

New customers acquired during the period do not belong in either numerator. These are retention measures, not total growth measures.

NRR, NDR, MRR, and ARR: keep the vocabulary straight

Retention discussions get muddied by interchangeable shorthand. Net dollar retention (NDR) is the same calculation as net revenue retention under a different label — vendors, investors, and term sheets use the two names for one metric. A buyer who wants to calculate NRR independently should rebuild it from customer-level records, not accept a vendor dashboard figure at face value.

Treat NDR and NRR as synonyms and confirm which formula sits behind either label before comparing figures across sellers.

The starting revenue base can be measured two ways. Monthly recurring revenue (MRR) is the current month's committed recurring revenue; annual recurring revenue (ARR) is that same figure annualized. A retention rate built on MRR can diverge from one built on ARR for a business with meaningful mid-year contract changes, so confirm which base a reported GRR or NRR actually used before treating it as a SaaS metric comparable to another company's.

None of these labels measures growth. Retention rate, gross revenue retention, and net retention describe what happened among existing customers in one cohort — not total revenue, and not how many new customers a company had to acquire to replace what it lost. Adding customers is a separate exercise, funded by sales and marketing, that these formulas deliberately exclude.

Why buyers should begin with GRR

GRR cannot exceed 100% under the standard definition because expansion is excluded. It exposes how much of the starting base is leaking away.

Imagine a clearly labeled hypothetical cohort that starts with $100,000 of annualized recurring revenue (ARR). During the period, $15,000 churns and $5,000 contracts. GRR is 80%. If remaining customers add $30,000 through upgrades, NRR is 110%.

The 110% NRR looks strong, but the company still lost 20% of its original base revenue. Expansion covered the leak. A buyer needs to know whether that expansion is repeatable and whether losing customers are concentrated in a product, segment, or acquisition channel.

This is why GRR is the retention floor. It is harder to decorate with a small number of large upgrades.

GRR is sometimes labeled gross retention or gross dollar retention. Confirm the seller used the standard definition — some vendors quietly exclude downgrades or net out credits, which inflates the figure and hides part of the customer retention story.

Why NRR still matters

NRR captures an important property of subscription economics: existing customers can become more valuable over time. Seats expand, usage grows, products cross-sell, and prices change. Revenue from existing customers, not new logos, drives the expansion side of the formula.

That growth can reduce dependence on constantly replacing churned customers. But it is not free. Buyers should identify what produced expansion:

• Genuine product adoption
• Contracted seat or usage growth
• One-time migration or services charges incorrectly included
• Price increases
• Currency movements
• Acquisition-related customer changes
• A few large enterprise upgrades

If one customer created most expansion, reported NRR may say more about concentration than broad product strength.

Strong NRR also changes unit economics elsewhere in the business. A company that can grow revenue from its existing base needs to acquire new customers only to add incremental total revenue, not to replace what churned — the same dynamic that improves customer lifetime value math across the SaaS metrics stack.

A seller who can document what actually helped improve net revenue retention — genuine product adoption versus a one-time price action — gives a buyer a clearer signal than the headline number alone.

How to calculate net revenue retention from one cohort

Retention comparisons fail when definitions drift. Use customer-level recurring revenue at the start and end of one consistent period. Keep the currency, customer population, products, and treatment of acquisitions constant.

SaaS Capital's 2025 benchmark research follows a year-over-year revenue cohort. Its GRR method caps each customer's ending revenue at the starting amount, preventing upsell from entering the gross measure — so the percentage of revenue retained reflects losses only, with expansion held out entirely.

A buyer should rebuild the seller's numbers and document:

1. Starting cohort date and revenue. 2. Which products count as recurring. 3. Treatment of pauses, credits, refunds, and failed payments. 4. Treatment of usage-based customers. 5. Whether price increases count as expansion. 6. Currency conversion policy. 7. Customers acquired or disposed through M&A.

Small changes can materially alter the result, especially with a small customer base.

What is a good net revenue retention rate?

There is no credible universal retention threshold for every SaaS company. Contract size, market, product type, maturity, customer segment, and billing model all affect normal behavior. Retention rate and churn rate are two sides of the same cohort, so a benchmark quoted for one industry's SaaS companies does not transfer cleanly to a business with a different churn rate, sales motion, or revenue growth profile.

SaaS Capital's 2025 private B2B study reports retention by annual contract value. For businesses with $25,000 to $50,000 ACV, it reported median NRR of 102% and an upper quartile of 111%. The research also found higher NRR associated with higher growth across its surveyed population.

That does not mean a consumer app should be judged against an enterprise-software cohort. Higher-price products often have implementation, sales, support, and switching patterns that make them structurally different.

ChartMogul's 2025 AI retention analysis makes the segmentation problem visible. Its dataset of roughly 3,500 software companies reported median NRR of 82% for B2B SaaS, 49% for B2C, and 48% for AI-native companies that had reached at least $250,000 ARR. The report also found sharply different retention across AI price bands.

Use those figures as research context, not a mechanical grading curve. The seller's own multi-period cohorts matter more than a mismatched industry headline.

Public-company NRR is useful—and different

In an SEC filing for the period ended October 31, 2025, Snowflake reported NRR of 125% — a high NRR reading it connected to increased customer workload consumption among existing customers. It separately warned that deferred revenue was not necessarily a meaningful indicator of when future revenue would be recognized.

That is a useful operating example, not a comparable valuation for a small private SaaS. Snowflake's scale, consumption model, customer mix, reporting controls, and capital-market context differ materially.

The transferable lesson is narrower: define the metric, explain its economic mechanism, and keep it separate from accounting measures it does not replace.

Retention quality checks buyers should run

These retention metrics work as a set, not in isolation. A retention formula that quietly nets out credits, or an NRR calculation that ignores segment mix, can understate revenue loss or overstate the revenue generated by real expansion.

Customer-count retention

Revenue retention can improve while many small customers leave and a few large accounts expand. Compare logo retention and customer counts with revenue retention — this is the customer retention side of the ledger, and a healthy revenue number can sit on top of an unhealthy customer-count trend.

Segment retention

Break out enterprise, SMB, self-serve, geography, product, plan, and acquisition channel. A blended average can conceal a failing segment.

Cohort maturity

Young cohorts may not have reached their main renewal or cancellation point. Compare mature cohorts at equivalent ages.

Margin retention

Expansion revenue with heavy support, cloud, payment, or model costs may add less profit than its ARR suggests. Measure retained gross profit where records permit.

Concentration

Calculate how NRR changes when the largest expansion customer is removed. This is a stress test, not a replacement metric.

Contract and transfer risk

Verify assignment, change-of-control, termination, pricing, and renewal provisions. Historical retention cannot guarantee post-acquisition retention.

How retention affects RSW's valuation reasoning

RSW does not add a fixed multiple when NRR crosses 100% or subtract one when GRR falls below a benchmark. Retention evidence changes confidence in the sustainability of earnings.

Strong GRR across mature, diversified cohorts suggests the base revenue is durable. Healthy NRR built on broad product adoption can support a stronger growth case. Weak GRR, expansion concentration, inconsistent definitions, or unverified exports increase uncertainty.

The valuation still begins with normalized financial performance and a suitable method. Retention explains the quality and risk behind the result.

Buyer checklist

Request the records needed to rebuild net revenue retention and gross revenue retention independently, not just the headline percentages:

• Customer-level starting and ending recurring revenue
• GRR, NRR, and logo retention under written definitions
• Monthly and annual cohorts
• Segment and price-band retention
• Expansion split among seats, usage, price, and cross-sell
• Churn rate and downgrade records by reason
• Gross margin by product or segment
• Top-customer concentration and contract terms
• A reconciliation to billing and accounting records

If the seller cannot reproduce the metrics from source data, treat them as marketing claims rather than diligence evidence.

FAQs

Can NRR exceed 100%?

Yes. Expansion from the starting customer cohort can exceed churn and contraction. GRR should not exceed 100% because it excludes expansion.

Is 100% NRR automatically good?

No. It can mask weak GRR if expansion from a few customers offsets substantial losses. Compare both metrics, concentration, margin, and customer-count retention.

Should new customers be included?

No. Standard NRR and GRR follow customers present at the start of the period. New-customer revenue belongs in total growth analysis.

Which benchmark should a buyer use?

Use a cohort that resembles the company's ACV, market, product, customer segment, and maturity. The company's historical cohort trend remains the most relevant comparison.

Is NRR the same as net dollar retention?

Yes. Net dollar retention (NDR) and net revenue retention describe the identical calculation — starting recurring revenue, plus expansion, minus churn and contraction, divided by the starting figure. Different vendors and investors default to different labels; the formula does not change.

What does an NRR above 120% mean?

It means expansion revenue from the existing customer cohort exceeded churn and contraction by a wide margin during that period. It says nothing on its own about GRR, customer-count retention, or concentration — verify what produced the expansion before treating a high number as evidence of broad-based product strength.

Continue through the RSW silos

Start with the SaaS valuation pillar and SaaS MRR valuation. Then reconcile churn's effect on value, LTV to CAC, and B2B versus B2C SaaS valuation. The website valuation multiples guide shows where retention quality enters the broader pricing framework.

Estimate the range, then rebuild retention

Use the Real Site Worth website value calculator for an automated starting range. Before buying or selling a SaaS business, rebuild GRR and NRR from customer-level records and connect them to contracts, margins, and concentration.