NRR, short for net revenue retention, is the SaaS metric that measures how much recurring revenue a fixed group of existing customers generates at the end of a period compared with the start, after adding expansion and subtracting contraction and churn. It is one of the central SaaS metrics because it shows, in a single percentage, whether the customer base grows or shrinks on its own, without any new customers.
In SaaS & startups, NRR is closely watched by investors and boards. A value above 100% means that upsells, extra seats and higher usage more than offset lost revenue, so the business grows even if sales stopped. A value below 100% means the existing base is leaking revenue that new sales must replace first. The metric is also called net dollar retention.
How NRR is calculated
NRR is calculated by taking the recurring revenue of a customer cohort at the start of a period, usually the annual recurring revenue twelve months ago, and comparing it with the revenue from exactly the same customers today. Revenue from customers acquired during the period is excluded.
The formula is: NRR = (starting revenue + expansion - contraction - churned revenue) / starting revenue. Most companies report a trailing twelve-month figure based on ARR, while some calculate it monthly from MRR and annualize it, which can produce different results.
The components of the formula are summarized below.
| Component | What it includes | Effect on NRR |
|---|---|---|
| Starting revenue | Recurring revenue of the cohort at period start | Denominator and baseline |
| Expansion | Upsells, extra seats, higher usage, add-ons, price increases | Raises NRR |
| Contraction | Downgrades, fewer seats, lower usage, discounts | Lowers NRR |
| Churned revenue | Revenue from customers who canceled entirely | Lowers NRR |
| New customers | Accounts acquired during the period | Excluded from the calculation |
Interpreting NRR values
NRR values are interpreted against 100%, the break-even point at which expansion exactly offsets losses. What counts as good depends heavily on the customer segment and pricing model, because small customers churn more often and have less room to expand.
| NRR range | Interpretation | Typical context |
|---|---|---|
| Below 90% | Base shrinks noticeably every year | Small-business products with high churn |
| 90-100% | Losses roughly match expansion | Products with little upsell potential |
| 100-110% | Healthy organic growth from the base | Many mid-market SaaS companies |
| 110-120% | Strong expansion | Seat or usage growth inside accounts |
| Above 120% | Exceptional expansion | Usage-based infrastructure and enterprise software |
These ranges are commonly used reference points, not fixed standards. A product sold to small businesses may be healthy at 95%, while an enterprise data platform at that level would raise concern.
NRR compared with related SaaS metrics
NRR is often confused with metrics that look similar but answer different questions about retention and growth. The comparison below separates them.
| Metric | Question it answers | Key difference from NRR |
|---|---|---|
| NRR | Does the existing base grow on its own | Includes expansion, can exceed 100% |
| Gross revenue retention | How much revenue is kept before expansion | Ignores expansion, capped at 100% |
| Logo churn | How many customers leave | Counts accounts, not revenue |
| Churn rate | How quickly customers or revenue are lost | Measures losses only |
| Expansion MRR | How much extra monthly revenue existing customers add | Measures gains only |
| SaaS quick ratio | How fast revenue grows relative to losses | Includes new customer revenue |
NRR and gross revenue retention should always be read together. A high NRR combined with a low GRR means that a few expanding accounts hide significant churn elsewhere, a pattern that becomes risky if those large accounts slow down.
What drives NRR and how to improve it
NRR is driven by two levers: reducing losses and increasing expansion. Pricing design determines how much expansion is even possible, while product value and customer success determine how much revenue is retained.
| Lever | Mechanism | Example action |
|---|---|---|
| Seat-based pricing | Revenue grows as teams grow | Encourage invitations of colleagues |
| Usage-based pricing | Revenue follows customer consumption | Charge per transaction or data volume |
| Tier upgrades | Customers move to higher plans | Place advanced features in upper tiers |
| Customer success | Prevents churn and finds expansion | Regular reviews with key accounts |
| Annual contracts | Reduce cancellation opportunities | Offer a discount for yearly billing |
Price increases also raise NRR, but expansion driven only by price is less durable than expansion driven by real growth in usage, because it can trigger downgrades and churn at the next renewal.
Pitfalls in measuring and reporting NRR
Because there is no single mandatory definition, NRR figures from different companies are often not directly comparable. Investors therefore ask how the metric was calculated before comparing values.
- Including new customers in the ending revenue inflates the result and is a calculation error.
- Monthly annualization can exaggerate NRR when early expansion is unusually strong.
- Counting one-time fees such as setup or services distorts a recurring revenue metric.
- Small cohorts produce volatile values in which one large account dominates.
- Usage-based revenue can fluctuate seasonally, so a single quarter may mislead.
A sound reporting practice states the cohort definition, the time window and the revenue basis alongside the percentage. Showing NRR by customer segment, for example small business, mid-market and enterprise, also prevents a strong segment from hiding a weak one. Tracking the figure quarter by quarter on a trailing twelve-month basis reveals trends early while smoothing out short-term noise.
In simple terms
NRR answers one question. If a company stopped finding new customers, would its income from current customers grow or shrink? It compares what a group of customers paid a year ago with what the same customers pay today.
Some customers spend more, some spend less and some leave. Above 100% means the group now pays more in total. Below 100% means it pays less.
A practical example
A project management software company starts the year with a cohort of customers paying 1,000,000 dollars in annual recurring revenue. During the year, existing customers add seats and upgrade plans worth 180,000 dollars, downgrades reduce revenue by 40,000 dollars, and cancellations remove 60,000 dollars.
The same customers now pay 1,080,000 dollars, so NRR is 108%. Gross revenue retention, which ignores the expansion, is 900,000 divided by 1,000,000, or 90%.
The leadership team concludes that expansion is healthy but that churn is too high. It invests in onboarding for small accounts, and the following year GRR rises to 93% and NRR to 112% with a similar level of expansion, showing how reducing losses lifts both metrics.