If you have one metric on the first slide of a SaaS deck, it should be net revenue retention. It compresses churn, contraction and expansion into a single number describing what happens to your existing revenue base over a year without any new sales.
Investors weight it heavily because it is very hard to fake and very hard to fix quickly. Growth can be bought with sales spend. NRR cannot.
#The calculation
NRR = (Starting MRR + Expansion − Contraction − Churn) / Starting MRRCritically, new customers are excluded. You are measuring a fixed cohort of revenue that existed at the start of the period and asking what it is worth now.
The related measure, gross revenue retention, omits expansion:
GRR = (Starting MRR − Contraction − Churn) / Starting MRRGRR can never exceed 100%. It tells you how leaky the bucket is. NRR tells you whether expansion is filling it faster than it leaks.
#What the benchmarks mean
- Below 90% — structurally difficult. You must acquire aggressively just to stand still, and every dollar of acquisition spend is partly replacement rather than growth.
- 90% to 100% — typical of SMB and self-serve products, where churn is naturally higher. Workable, but growth is entirely dependent on new sales.
- 100% to 110% — solid. Expansion offsets churn.
- 110% to 125% — strong. The base compounds on its own.
- Above 125% — exceptional, usually usage-based pricing or a product embedded deeply in a growing workflow.
The compounding effect is what makes the difference so stark. A business at 130% NRR roughly doubles its existing base every three years with no new customers. A business at 90% loses a third of it.
#Why it predicts valuation
Three reasons, and they reinforce each other.
Revenue quality. High NRR means revenue is durable and expanding, so a forecast built on it is more reliable.
Capital efficiency. If the base grows on its own, less of your sales spend goes to replacement, so each dollar of acquisition buys more net growth.
Optionality. A company at 125% NRR can slow acquisition spend during a downturn and still grow. One at 90% cannot stop without shrinking.
Public SaaS valuation multiples correlate more tightly with NRR than with growth rate alone, which is why diligence asks for it by cohort rather than in aggregate.
#The measurement mistakes
Including new customers. The most common error. New logos belong in growth, not retention. Including them produces a flattering number that means nothing.
Not defining the cohort. NRR is a twelve-month measure on a defined starting cohort. A monthly figure naively annualised will overstate a good month and cause panic on a bad one.
Treating downgrades as churn. A customer moving from Enterprise to Pro is contraction. Recording it as churn overstates churn and understates the base.
Mixing segments. Self-serve and enterprise have entirely different retention profiles. A blended 103% might be enterprise at 118% and self-serve at 84% — two completely different problems disguised as one acceptable number. Run the SaaS metrics calculator separately per segment.
#How to actually move it
Expansion pricing. The structural fix. If your pricing scales with a metric that grows as the customer succeeds — seats, usage, volume, revenue processed — NRR rises automatically as customers grow. Flat per-account pricing caps NRR at 100% by construction.
Onboarding. Most churn is decided in the first ninety days. Customers who reach a defined activation milestone retain dramatically better than those who do not. Instrument the milestone, then measure the proportion reaching it.
Multi-product. Customers using two or more products churn at a fraction of the rate of single-product customers, in essentially every dataset. Cross-sell is a retention strategy before it is a revenue strategy.
Annual contracts. They do not fix a product problem, but they convert twelve monthly cancellation decisions into one, which materially reduces churn events.
Churn diagnosis. Distinguish involuntary churn — failed cards, expired payment methods — from voluntary. Involuntary churn is commonly 20% to 40% of total and is fixed with dunning and card updater services, which is engineering work rather than product work.
#The honest caveat
NRR flatters businesses with usage-based pricing during a growth period and punishes them during a contraction, since usage falls with customer activity. Read it alongside gross retention: a company at 115% NRR and 82% GRR is losing a lot of customers and hiding it behind expansion from the survivors. That is a fragile position that looks strong in a single number.
Frequently asked questions
What is a good net revenue retention rate?
Above 110% is strong for B2B SaaS and above 125% is exceptional. SMB and self-serve products typically run 90% to 105% because churn is structurally higher. Always compare against businesses with a similar customer profile.
What is the difference between NRR and NDR?
They are the same metric. Net revenue retention and net dollar retention are used interchangeably. Some companies compute it on ARR rather than MRR, which gives the same result if calculated consistently.
Can NRR be above 100% with high churn?
Yes, if expansion from remaining customers exceeds the revenue lost. That is why gross revenue retention should always be reported alongside it — high NRR masking low GRR is a fragile position.