Net Revenue Retention: The Expansion-Revenue Playbook

Net revenue retention (NRR) measures whether the revenue you keep from existing customers is growing or shrinking each period, counting expansion against downgrades and cancellations. Crossing 100% is the line where the base starts funding its own growth.

Across the subscription businesses we’ve helped, the companies that consistently cross it hold a tighter grip on the revenue they already have while triggering expansion. Do both and your existing base compounds, with no new acquisition needed to cover the losses.

Key takeaways

• Track net revenue retention above 100% to confirm the existing base self-funds growth.
• Grow expansion revenue, the only NRR input you can add to.
• Close your gross revenue retention gap before you scale any upsell motion.
• Read a high rate over falling gross retention as expansion hiding churn.
• Run the three playbook stages in order, Defend then Diagnose then Drive.
• Companies with strong NRR grow 2.5x faster than those without it.

What is net revenue retention (NRR)?

Net revenue retention is the share of recurring revenue you keep from an existing customer cohort after expansions, contractions, and cancellations. Above 100%, that cohort generates more revenue than it started with.

Because it counts no new customers, it isolates one question. If you signed nobody new this period, would the existing base's revenue rise or fall?

To calculate it, take the cohort's starting monthly recurring revenue (MRR), add what they expanded, subtract what they downgraded and canceled, then divide by where they began:

NRR = (Starting MRR + Expansion MRR − Contraction MRR − Churned MRR) ÷ Starting MRR × 100

Four inputs feed that calculation, and each measures a different kind of revenue movement inside the base:

• Starting MRR: the cohort's recurring revenue at the start of the period.
• Expansion MRR: what those customers add through upgrades, seat adds, and usage growth.
• Contraction MRR: what they take back when they downgrade or drop seats.
• Churned MRR: what you lose when they cancel outright.

Only expansion adds to the base. The other three subtract, which is why expansion is the only input that can lift NRR while contraction and churn drag it down. The full derivation and a worked example sit in the NRR glossary entry, so this guide focuses on what moves the number.

NRR leaves new-logo revenue out by design, so a low number isn't the same as a stalling business. Say a business shows 80% NRR but 50% monthly new-logo growth. Still growing fast.

Does NRR include new customers?

No, NRR counts only the customers who were already in the cohort when the period began. Anyone you sign during the period drops out of the math, which is what lets NRR isolate base health from new-sales growth.

What is the difference between NRR and NDR?

Net revenue retention and net dollar retention (NDR) are the same metric on the same formula, and the only thing that differs is which audience uses which name.

Investors and public filings tend to say NDR, while SaaS teams and billing dashboards say NRR, as the net dollar retention breakdown explains.

Why most subscription businesses stall below 100% NRR

Most subscription businesses stall below 100% because they treat churn prevention and expansion as two separate programs. You lose revenue to churn faster than expansion can replace it, so the gain never shows up in the NRR ratio.

The upsell team books new expansion while the cancel page loses the same amount, and the net number barely moves.

The reason is sequence.

Gross revenue retention is the floor your expansion has to clear before it counts as growth. At 85% gross revenue retention (GRR), expansion has to produce 15 cents per starting dollar just to reach parity.

Only the cents above that lift NRR past 100%, so a team that scales upsells before working to reduce churn underneath them is paying to stand still.

That changes at the bottom of the pricing range. Consider a product under about $10 a month. It usually has no real expansion path, with no seats to add, no tier to climb, and no usage to meter. 

There, GRR and NRR are effectively the same number, and defending the base is the entire retention strategy. For everyone with room to expand, the gap between what you keep and 100% is the number to size first.

What high NRR actually signals

NRR above 100% is a compounding signal, and companies that hold it tend to outgrow the ones that don't. High Alpha's 2025 research found that SaaS companies with high NRR grow 2.5x faster than their low-NRR counterparts.

The mechanism is what a weak quarter does on each side of the line. Below 100%, the customers you have lose more than they add, so a slow acquisition month shrinks the base. 

High Alpha's 2025 research links this directly to slower growth. Above 100%, that same slow month only slows your growth, because the base keeps expanding while you fix the top of the funnel.

That decoupling matters more each year as acquisition costs rise.

David Skok's cohort math makes that compounding visible. Suppose a business runs 3% monthly churn on steady bookings, per Skok's model, and revenue climbs to roughly $140,000 before it flattens.

Run the same model at 3% monthly negative churn and revenue reaches about $450,000, still climbing. Same bookings and same churn rate, with the only difference being whether the base shrinks or grows into negative churn.

A high number still has to be read against the gross retention underneath it. When NRR is strong but GRR is sliding, expansion is covering heavy churn. The moment the upsell pipeline slows, the number corrects downward.

That masking case is worked through in the GRR vs NRR comparison.

What does NRR above 100% mean?

NRR above 100% means your existing customers reliably add more revenue than they drop, so the base grows on its own each period.

High Alpha's 2025 research marks that threshold as the inflection to a 2.5x growth advantage. Investors read a high figure as proof the product captures more value as customers grow. The team running the business reads it as room to let a weak acquisition quarter slow growth rather than reverse it.

The NRR Expansion Playbook: three stages

The NRR Expansion Playbook defends the base, diagnoses the NRR/GRR gap, then drives expansion, in that order because expansion only compounds once churn is contained. Drive expansion before you defend the base and you just produce revenue that replaces losses.

Each stage feeds the next:

1. Defend: raise gross revenue retention so expansion has less to replace.
2. Diagnose: read the gap between your NRR and GRR to find the binding constraint.
3. Drive: trigger the expansion lever that fits the constraint you found.

The sections below take each stage in turn, from defending the base up to the expansion offer.

Stage 1. Defend: raise the GRR floor

Defending gross revenue retention means cutting the cancellations and downgrades that drag it down. Every point of GRR you recover is a point of expansion you no longer have to spend replacing.

A save in a cancel flow converts churned MRR into kept MRR right inside the NRR formula, which moves GRR directly.

Picture 10% of subscribers canceling each month on $100,000 MRR, your save rate climbing from 18% to 28%, and GRR sitting at 85%. Those 10 extra saves keep about $1,000 of MRR, nudging GRR toward 86%.

The three levers that move GRR are recovering cancellations, saving downgrades, and rescuing failed payments. All three are worked through in the GRR Defense Playbook.

The lowest-friction save is a pause.

Recurly's 2025 report found that 25% of subscribers pause instead of canceling where the option exists, which keeps the account in the base. Recover the customers you can actually save, because a hard-to-leave flow that holds someone this month means fewer of them come back next year.

See how Churn.io defends the GRR floor by catching would-be cancels in the flow with the right offer at the moment of the decision.

Stage 2. Diagnose: read the NRR/GRR gap

Diagnosing the gap means reading your NRR and GRR side by side, because the distance between them tells you what's actually holding you back. One number alone hides whether the problem is too much churn, too little expansion, or both. The spread tells you.

A GRR below roughly 85% with a healthy gap says churn is the problem, so the work is back at Stage 1. The SaaS retention benchmarks break that threshold down by revenue stage. A strong GRR with a narrow gap says the base is solid but expansion is thin, so the work is the offer motion in Stage 3.

A wide gap that's widening, where NRR holds while GRR slides, is the masking case from above. It sends you back to defending the base before the upsell pipeline runs dry.

Reading that signal needs more than the headline rate, which is why a customer health score belongs in the diagnosis. It flags which accounts are drifting before the gap reaches the quarterly number.

The diagnosis only works when the two numbers can diverge, which rules out flat-rate products.

Run the gap over at least four quarters before acting on it. A single reading can't separate a steady gap from a widening one. The pitfall is treating a wide gap as automatic trouble. On a flat GRR it is just strong expansion, so the GRR direction underneath decides.

Stage 3. Drive: trigger expansion at the right moment

Driving expansion means firing the right offer on a behavioral product signal, so the upgrade reads as the fix to a constraint the customer already feels. An offer timed to a renewal date feels like a sale. An offer timed to a usage signal feels like service.

Three signals map to three moves:

• Plan-limit signal: an account sitting at its seat cap or usage quota across two billing cycles points to an upsell, a higher tier of what they already buy.
• Adjacent-job signal: a workflow or integration that reveals a second need beside the one your product serves points to a cross-sell, a separate module.
• Team-growth signal: new users provisioned or usage climbing month over month points to seat or usage expansion.

The build for these triggers runs through our Upsell Trigger Framework, the expansion revenue playbook, and for metered products, the usage-based pricing guide.

One rule sits above all three. An at-risk account, flagged by a falling health score, a support escalation, or dropping usage, gets a save offer rather than an upgrade prompt.

Selling to a customer who's already pulling away just confirms the doubt and speeds the exit.

Confirm the expansion held before you count it. An upgrade that reverses within a quarter inflates expansion MRR then books as churn. Track whether expanded accounts hold their new spend over the next two billing cycles.

Use the NRR calculator to model what each stage is worth before you build.

FAQ

What is a good NRR for SaaS?

A good NRR is 100% or higher at most stages, with 110% and above marking top-quartile performance. The target shifts with revenue stage and contract value, so an SMB product near parity and an enterprise product well above it can both be healthy.

How is NRR different from gross revenue retention (GRR)?

NRR and GRR run the same calculation, except NRR adds expansion back in, which is why NRR can exceed 100% while GRR never can. 

How often should NRR be measured?

Measure NRR on a rolling 12-month window with a fixed cohort, so each period compares like with like. A single-month snapshot swings too much with seasonal noise to act on.

Can NRR be too high?

NRR itself can't be too high, but its composition can hide a problem. The same high rate built on heavy churn drops fast if expansion stalls, while the same rate on a solid gross base holds.