Net revenue retention (NRR) is the single most revealing metric about the health of a recurring revenue business. It captures whether your existing customer base is growing or shrinking in value over time, independent of new customer acquisition. A company with strong NRR can grow its revenue even in a period of slow new logo acquisition. A company with weak NRR is running a leaky bucket regardless of how many new logos it adds.
Understanding how to calculate NRR correctly matters as much as the number itself. Most teams calculate it wrong, which makes the output unreliable for planning.
Step 1: Define the Inputs
NRR requires four data inputs for a defined time period. Use monthly recurring revenue (MRR) or annual recurring revenue (ARR) consistently. Do not mix them.
| Input | Definition |
|---|---|
| Starting MRR | MRR from a defined cohort or period at the start |
| Expansion MRR | New MRR added from existing customers via upsell or cross-sell |
| Contraction MRR | MRR lost from existing customers via downgrades |
| Churned MRR | MRR lost from customers who cancelled entirely |
Step 2: Apply the NRR Formula
``` NRR = (Starting MRR + Expansion MRR - Contraction MRR - Churned MRR) / Starting MRR ```
Express the result as a percentage. An NRR of 105% means the cohort or period produced 5% more revenue at the end of the period than it generated at the start, after accounting for all losses and gains within the existing customer base.
Worked example: If you start a quarter with customers generating a combined $400,000 in MRR, add $60,000 in expansion from upgrades, lose $15,000 to downgrades, and lose $25,000 to cancellations:``` NRR = ($400,000 + $60,000 - $15,000 - $25,000) / $400,000 NRR = $420,000 / $400,000 NRR = 105% ```
Step 3: Choose Between Cohort and Point-in-Time Methods
The formula is the same, but which customers count as "starting MRR" differs between the two approaches.
Cohort method: Fix a specific group of customers, typically those who started in a given month or quarter, and track that group over 12 months. This is the most accurate picture of how customer relationships evolve. It is also more work to maintain because you need to track individual accounts over time. Point-in-time method: Compare all recurring revenue across two snapshots of time, typically the same month in the prior year versus the current month. This is easier to calculate from a standard CRM or billing extract. It is the version most boards and investors request. It is also susceptible to distortion if the customer mix shifts significantly between periods.Use cohort NRR for internal analysis and product decisions. Use point-in-time NRR for external reporting where a consistent calculation is more important than perfect cohort tracking.
Step 4: Pair NRR with Gross Revenue Retention
NRR above 100% can coexist with significant logo churn. Gross revenue retention (GRR) exposes this.
``` GRR = (Starting MRR - Contraction MRR - Churned MRR) / Starting MRR ```
GRR never exceeds 100%. It shows how much of your starting revenue base you retained, excluding any expansion. Using the same example: GRR would be ($400,000 - $15,000 - $25,000) / $400,000 = 90%.
Reading NRR (105%) alongside GRR (90%) together shows that the business is expanding revenue per customer but losing 10% of its base revenue to churn and downgrades. That gap has implications for long-term durability, customer concentration risk, and the true cost of growth.
Step 5: Decompose NRR into Its Drivers
NRR as a single number is a lagging indicator. The analysis becomes useful when you decompose it into the four inputs separately.
| Component | What movement tells you |
|---|---|
| Rising expansion | Customers are finding more value, upsell motion is working |
| Rising contraction | Customers are reducing scope, possibly before churning |
| Rising churn | Customer success or product-market fit issue |
| Flat or falling expansion with stable churn | Growth plateau within existing accounts |
Common Mistakes
Including new logo MRR in expansion. New customers are not expansions. Including them inflates NRR and makes it look like existing customers are growing when they are not. Using different populations for starting and ending MRR. If the set of customers in your "starting" population is different from the set you measure at period end, the calculation is not comparing like to like. Calculating NRR on a monthly basis and calling it annual NRR. Monthly and annual NRR use the same formula but produce numbers that are not comparable. Be explicit about the measurement period. Ignoring GRR. Teams that report only NRR can hide a deteriorating retention foundation behind strong expansion. Report both.Frequently Asked Questions
What is a strong NRR for a B2B SaaS company?
NRR above 100% means expansion revenue is outpacing churn and contraction. The precise level that qualifies as strong varies by growth stage, motion, and segment. What matters more than any external reference is whether your NRR is improving, stable, or declining over time, and what is driving the movement.Can NRR be above 100% if you are losing customers?
Yes. NRR measures revenue dollars, not logo count. If your churning customers are small and your expanding customers are large, NRR can exceed 100% even with logo churn. This is why NRR and gross revenue retention (which excludes expansion) must be read together.What is the difference between cohort NRR and point-in-time NRR?
Cohort NRR tracks a fixed group of customers from a starting period and measures how their revenue has changed over time. Point-in-time NRR compares total recurring revenue across two periods without fixing the customer population. Cohort NRR is more accurate for understanding customer behavior. Point-in-time NRR is faster to calculate and used in most investor reporting.How does NRR differ from gross revenue retention?
Gross revenue retention (GRR) measures how much of your starting revenue you kept, excluding any expansion. NRR includes expansion revenue in the numerator. GRR can never exceed 100%. NRR can. A high NRR with low GRR signals that you are growing revenue from existing accounts but losing a meaningful share of customers.For related concepts, see net revenue retention, gross revenue retention, and expansion revenue.
Frequently Asked Questions
What is a strong NRR for a B2B SaaS company?
NRR above 100% means expansion revenue is outpacing churn and contraction. The precise level that qualifies as strong varies by growth stage, motion, and segment. What matters more than any external reference is whether your NRR is improving, stable, or declining over time, and what is driving the movement.
Can NRR be above 100% if you are losing customers?
Yes. NRR measures revenue dollars, not logo count. If your churning customers are small and your expanding customers are large, NRR can exceed 100% even with logo churn. This is why NRR and gross revenue retention (which excludes expansion) must be read together.
What is the difference between cohort NRR and point-in-time NRR?
Cohort NRR tracks a fixed group of customers from a starting period and measures how their revenue has changed over time. Point-in-time NRR compares total recurring revenue across two periods without fixing the customer population. Cohort NRR is more accurate for understanding customer behavior. Point-in-time NRR is faster to calculate and used in most investor reporting.
How does NRR differ from gross revenue retention?
Gross revenue retention (GRR) measures how much of your starting revenue you kept, excluding any expansion. NRR includes expansion revenue in the numerator. GRR can never exceed 100%. NRR can. A high NRR with low GRR signals that you are growing revenue from existing accounts but losing a meaningful share of customers.
See how ORM turns these insights into action
ORM builds custom revenue forecast models for B2B SaaS companies. Not dashboards. Prescriptive analytics that tell you what to do next.
Schedule a Demo