Net Revenue Retention: The SaaS Metric That Predicts Growth

This guide covers the NRR formula, how it differs from gross retention, a fully worked cohort example, benchmarks by segment, the mistakes that inflate the number, and five concrete levers to raise it.

What Is Net Revenue Retention?

Net revenue retention measures the percentage of recurring revenue retained from an existing customer cohort over a period, after accounting for churn, downgrades, upgrades, upsells, and cross-sells — but excluding revenue from new customers. It answers: is our installed base expanding or contracting?

The metric matters because it is the purest signal of product-market fit at scale. New logos can be bought with marketing spend; expansion revenue cannot — it must be earned through delivered value. When customers voluntarily spend more over time, it proves the product is becoming more embedded, not less. That is why NRR correlates so strongly with valuation multiples: public SaaS companies above 120% NRR have historically traded at roughly double the revenue multiples of those below 110%.

NRR also determines how much growth must come from acquisition. At 120% NRR, a company targeting 40% overall growth only needs 20 points from new logos. At 90% NRR, the same target requires 50 points of new-logo growth — a far more expensive and fragile motion.

The Net Revenue Retention Formula

NRR = [(Starting MRR − Churned MRR − Downgraded MRR + Expansion MRR) ÷ Starting MRR] × 100

Each component:

  • Starting MRR. The monthly recurring revenue of the cohort at the beginning of the period. NRR is always measured on a fixed cohort — typically all customers at the start of the quarter or year — so new customers acquired during the period are excluded.
  • Churned MRR. Revenue lost from customers who cancelled entirely.
  • Downgraded MRR. Revenue lost from customers who moved to cheaper plans or reduced seats. Often overlooked, downgrades are partial churn and must be counted.
  • Expansion MRR. Revenue gained from the same cohort through upsells, cross-sells, seat additions, plan upgrades, and usage overages. This is what can push NRR above 100%.

The companion metric is gross revenue retention (GRR):

GRR = [(Starting MRR − Churned MRR − Downgraded MRR) ÷ Starting MRR] × 100

GRR excludes expansion, so it can never exceed 100%. It measures pure retention — how well the product holds what it has. NRR measures retention plus expansion. The gap between them reveals the expansion engine: a company at 92% GRR and 118% NRR loses 8% of its base but expands the survivors by 26 points.

Worked Example: A 110% NRR Cohort

Consider Corelink, a B2B SaaS company measuring NRR on its Q1 customer cohort over the following 12 months. Starting MRR: $500,000 across 250 customers.

Over the year, the cohort's revenue moves:

  • Churned (cancelled): $42,000 MRR lost
  • 18 SMB customers churn: $18,000
  • 4 mid-market customers churn: $24,000
  • Downgraded (fewer seats / cheaper plans): $15,000 MRR lost
  • Expansion:
  • Seat additions on existing accounts: $38,000
  • Plan upgrades (Pro → Enterprise): $45,000
  • Usage overages (API calls beyond plan): $24,000
  • Total expansion: $107,000

Step 1 — Compute gross revenue retention.

($500,000 − $42,000 − $15,000) ÷ $500,000 = $443,000 ÷ $500,000 = 88.6%

Corelink keeps 88.6% of its starting revenue before expansion — solid core retention with 11.4% leakage.

Step 2 — Compute net revenue retention.

($443,000 + $107,000) ÷ $500,000 = $550,000 ÷ $500,000 = 110.0%

Step 3 — Interpret.

The cohort now generates $50,000 more monthly revenue than it started with, despite losing $57,000 to churn and downgrades. Expansion of $107,000 — nearly double the leakage — more than compensates. At 110% NRR, Corelink's installed base grows 10% annually with zero new sales, and every new logo is pure incremental growth on top.

Note the composition insight: $83,000 of the $107,000 expansion (78%) comes from seat additions and plan upgrades — motions the product and customer success teams control. Only $24,000 comes from usage overages. If Corelink wants to push NRR toward 115%+, the highest-leverage investment is in the upgrade path, not in acquiring more overage-prone accounts.

Benchmarks: What Is a Good NRR?

NRR benchmarks by segment:

  • SMB SaaS: 90–105% is typical; above 100% is strong for high-volume, low-touch products. SMB cohorts churn more, so expansion must work harder.
  • Mid-market SaaS: 105–115% is the healthy range. Best-in-class mid-market companies sustain 115–125%.
  • Enterprise SaaS: 110–130%+. Enterprise NRR benchmarks are the highest in SaaS — large accounts expand through multi-year rollouts, divisional land-and-expand, and usage growth. Below 110% in enterprise suggests weak expansion motion.
  • Product-led growth (PLG): 105–125%, with the top quartile exceeding 120% driven by viral seat expansion and usage-based pricing.

General rules used by investors:

  • Below 100%: the installed base is shrinking. Growth depends entirely on new logos — expensive and fragile. Investigate churn and expansion separately to find the leak.
  • 100–110%: acceptable; the base holds roughly steady. Common for SMB-heavy businesses.
  • 110–120%: healthy. The expansion engine is working and new-logo growth compounds on top.
  • Above 120%: elite. Typical of best-in-class enterprise and PLG companies with strong land-and-expand motions.

For gross revenue retention, the benchmarks are tighter: above 90% annually is healthy for most segments, above 95% is excellent, and enterprise GRR above 90% is effectively table stakes.

5 Common NRR Mistakes

1. Including new-customer revenue. NRR is strictly a cohort metric. Blending in new logos inflates the number and hides whether the existing base is expanding or contracting. This is the single most common error.

2. Ignoring downgrades. A customer moving from $2,000 to $800/month is $1,200 of lost revenue — partial churn. Companies that count only full cancellations systematically overstate both GRR and NRR.

3. Measuring over inconsistent periods. NRR compounds: 102% quarterly NRR annualizes to roughly 108%, not 102%. Always annualize (or state the period) before comparing to benchmarks, and use the same period quarter to quarter.

4. Letting one whale distort the cohort. A single large account doubling its spend can mask broad-based churn across hundreds of small accounts. Segment NRR by customer tier so a whale cannot hide a leaking SMB base.

5. Treating expansion as free. Expansion revenue has a cost: customer success headcount, support load, and sometimes discounted upgrade pricing. NRR above 120% driven by deeply discounted expansions may be less profitable than 112% at full price. Pair NRR with gross margin by cohort.

Calculate Your Own Net Revenue Retention

Enter your starting MRR, churned and downgraded revenue, and expansion in the free net revenue retention calculator to get your NRR and GRR — then diagnose the churn side with the churn rate calculator.

Educational content only — not financial advice.