How do you calculate net revenue retention for a B2B SaaS company in 2027?

A subscription business tries to explain a confusing quarter

Picture a mid-market SaaS company closing out Q2 with a board deck due Friday. The CFO reports that revenue grew 18% year over year, but the VP of Customer Success is worried — three of the company's ten largest accounts churned during the quarter, and two more downgraded their seat counts. How can both things be true at once? The answer lives in the difference between total revenue growth (which includes new-logo sales) and net revenue retention (which strips new logos out entirely and asks a narrower question: how did the customers we already had behave?).

This is exactly the scenario where NRR earns its keep as a metric. The company's new sales team closed enough new business to mask the leakage happening underneath, in the existing base. If the board only sees the 18% top-line number, they will assume the business is healthy in every dimension. If they see NRR sitting at 91%, they learn something the blended number hides: the existing customer base is shrinking, and the company is running hard just to stand still. Calculating NRR separately from new-business bookings is how a company catches this kind of structural problem before it compounds. A business that depends entirely on new-logo acquisition to offset base erosion has a much more fragile growth engine than one where the existing base compounds on its own, and the gap only shows up once you isolate the cohort correctly.

How the NRR mechanism actually works

The calculation only works if you are disciplined about what counts as "the same cohort." Start by choosing a period — monthly or annual — and a revenue base, either MRR or ARR; the formula is identical either way, only the units change. Lock the cohort as of day one of the period: every customer paying you on that date, and only those customers, are eligible to move the number for the rest of the window.

From there, three movements get tracked against that fixed group. Expansion captures existing customers paying more — upsells to a higher tier, additional seats, cross-sells of a new module, or usage growth under a consumption model. Contraction captures existing customers paying less — a downgrade to a cheaper plan or a reduction in seat count, without a full cancellation. Churn captures customers who cancel entirely and drop to zero. Critically, revenue from any customer who signed a first contract after the cohort was locked never enters the numerator or denominator — including it is the single most common way this metric gets miscalculated, because it mechanically inflates the result and hides real churn behind new-logo bookings.

Once the mechanism is set up correctly, the calculation itself takes minutes each period — the hard part is holding the cohort boundary and the movement definitions constant so the trend is comparable quarter over quarter.

Real numbers, ranges, and benchmarks

A worked example makes the arithmetic concrete. Suppose a cohort starts the year with $1,000,000 in ARR. Over the year, existing customers add $200,000 in expansion revenue through upsells and seat growth, reduce spend by $50,000 through contraction, and fully churn $80,000 of the starting base. The numerator is $1,000,000 + $200,000 − $50,000 − $80,000 = $1,070,000. Divide by the $1,000,000 starting base and the company's NRR for the year is 107% — the existing base grew 7% with zero new customers added.

Segment-level ranges vary widely, and quoting one universal "good" number is misleading because pricing model, customer size, and go-to-market motion all shift what is achievable. Enterprise-heavy B2B SaaS companies with land-and-expand motions frequently see NRR in the 110-130% range, because large accounts have more seats and use cases to grow into over time. Mid-market companies more commonly land in the 100-115% range. SMB-heavy companies, where individual accounts are small and churn more easily, often sit below 100% even when the product is healthy, because the base simply turns over faster. Usage-based and AI-metered companies can show much higher swings in either direction — a single large customer's consumption spike can move segment-level NRR by ten points or more in a single month, which is why these companies increasingly report on a trailing three-month or rolling twelve-month basis rather than a single snapshot.

Gross revenue retention (GRR) is the number to calculate alongside NRR, because it puts a floor under the story. GRR uses the same starting base but only subtracts contraction and churn — it never credits expansion, so it is mathematically capped at 100%. GRR in the 90-95% range is generally considered healthy for a B2B SaaS company; anything meaningfully below 85% signals a product or onboarding problem regardless of how strong NRR looks. The gap between GRR and NRR is itself a benchmark worth tracking: a wide gap (say, GRR at 88% and NRR at 118%) means a handful of expanding accounts are masking real churn elsewhere in the base — a concentration risk that a single blended metric would never surface.

Trade-offs and alternatives in how companies calculate it

There is no single "correct" way to calculate NRR once a company moves past pure per-seat subscription pricing, and the choices involve real trade-offs. The first choice is period length. A monthly snapshot is the most granular and catches problems fastest, but for a consumption-priced company it is also the noisiest — a customer's usage might spike for a seasonal reason in one month and normalize the next, showing up as expansion then contraction with no real change in the underlying relationship. The alternative is a rolling three-month average or a trailing-twelve-month comparison, which smooths that noise at the cost of slower problem detection — a real churn signal takes longer to show up clearly.

The second choice is what counts as "revenue" in a usage-based or AI-metered contract. A company can report NRR on total consumption revenue including on-demand overage, which is more complete but more volatile, or it can separate committed contract revenue (which behaves like a traditional subscription) from variable overage, reporting NRR primarily on the committed base while tracking overage separately as a supplementary signal. The committed-only approach produces a calmer, more comparable trend line; the total-consumption approach produces a number that moves with real customer usage but is harder to forecast from.

A third trade-off is constant-currency versus reported-currency calculation. For any company with a meaningfully global base, foreign-exchange movement can add or subtract several points of apparent expansion or contraction that have nothing to do with customer behavior. Holding FX rates fixed at the start-of-period rate isolates the real signal but requires extra bookkeeping each period; reporting in raw currency is simpler but noisier and can mislead a board that reads currency swings as customer health changes. There is no universally "right" answer among these trade-offs — the right choice depends on how volatile the company's revenue base actually is — but whichever approach is chosen, it has to be applied consistently, because switching methods between quarters makes the trend line meaningless.

Common pitfalls and how to avoid them

The most common pitfall, repeated across nearly every company learning to calculate this metric for the first time, is letting new-logo revenue leak into the cohort. If a customer signed their first contract mid-period, their revenue must never appear in either the numerator or the denominator — including it mechanically pushes NRR above where it actually belongs and hides real churn happening elsewhere in the base. Any time an NRR figure looks unexpectedly strong, the first audit step is confirming the cohort boundary was respected.

A second pitfall is treating a discount roll-off as expansion. When an introductory or promotional discount expires and a customer reverts to list price, revenue rises on paper, but the customer did not actually buy more — this should be flagged separately, not folded into genuine expansion, or the metric will overstate how much customers are voluntarily growing with the product.

A third pitfall is inconsistent treatment of multi-year and prepaid contracts. A company must decide once whether to annualize total contract value or recognize it ratably period by period, and apply that same choice to every account — mixing methods within a single calculation produces a number that cannot be trusted or compared across periods.

A fourth pitfall is reactivations. A customer who churned and later signs a new contract should generally not be added back into their original cohort — doing so double-counts their revenue and distorts the churn figure from the period they originally left.

The fifth and most damaging pitfall is silently changing methodology between reporting periods — switching the smoothing window, redefining what counts as expansion, or changing the currency treatment — without restating prior periods to match. This is how a company loses finance and board trust in the metric entirely, because a trend line that was calculated two different ways is not actually a trend at all. The fix is simple: document the calculation method in writing, apply it identically every period, and if the method must change, restate history so the full trend remains honest.

Related questions

Can net revenue retention exceed 100%?

Yes. NRR credits expansion revenue, so when existing customers upgrade, add seats, or grow usage faster than others churn or downgrade, NRR rises above 100%. This "negative net churn" pattern is common in strong product-led and enterprise SaaS companies.

Should NRR be calculated using MRR or ARR?

Either works, as long as the choice is consistent. Monthly-billing SaaS typically uses MRR; annual-contract enterprise SaaS uses ARR. The formula does not change — only the revenue base and period length differ.

Does NRR include revenue from new customers?

No. NRR is calculated against a locked cohort, and new logos acquired during the period are excluded from both the numerator and denominator. Including them is the most common calculation error and artificially inflates the metric.

How often should a company calculate NRR?

Most companies calculate it monthly for internal tracking and quarterly for board and investor reporting. Usage-based businesses often prefer a trailing-twelve-month or rolling three-month average to smooth out consumption volatility rather than relying on a single-month snapshot.

What is the relationship between net churn and NRR?

They describe the same movements from opposite angles. NRR = 100% minus the net churn rate. When net churn is negative — meaning expansion outweighs losses — NRR exceeds 100%. Net revenue churn simply frames the identical calculation as a loss figure instead of a retention figure.

FAQ

What is a simple worked example of the NRR calculation? A cohort starts the year with $1,000,000 in ARR. Existing customers add $200,000 in expansion, reduce spend by $50,000 in contraction, and fully churn $80,000. The numerator is $1,000,000 + $200,000 − $50,000 − $80,000 = $1,070,000, divided by the $1,000,000 starting base equals 107% NRR — the base grew 7% before any new customers were added.

Why does NRR include expansion revenue but GRR does not? NRR is designed to capture the full revenue trajectory of the existing base, including its capacity to grow, so expansion is credited. GRR is designed to isolate pure leakage — what a company loses when nobody upsells — so expansion is deliberately excluded, which caps GRR at 100%. Reporting both together gives a company the complete retention story.

How do you handle a customer who downgrades and then upgrades within the same period? Net the movements for that customer against the same cohort. If they downgrade by $10,000 and later upgrade by $15,000 within the period, the net effect recorded is $5,000 of expansion. What matters is the start-to-end delta per account, applied the same way to every customer in the cohort.

Does NRR include one-time or non-recurring revenue?

No. NRR measures recurring revenue only — subscriptions, committed usage, and renewable contracts. One-time fees such as implementation, professional services, or setup charges are excluded because they are not repeatable revenue and would distort the retention signal.

How does usage-based pricing change how a company calculates NRR? The formula itself does not change, but the inputs need smoothing. Consumption revenue fluctuates for reasons unrelated to customer health, so companies often average revenue over a trailing three-month window or use a rolling twelve-month calculation. Separating committed contract revenue from variable overage keeps the metric from swinging on seasonal usage spikes.

Why can NRR and GRR tell opposite stories about the same company? A company can post strong NRR while GRR is weak when a few large accounts expand aggressively enough to mask meaningful churn spread across the broader base. The NRR looks healthy, but the revenue is concentrated in a handful of accounts and is fragile. Calculating and reporting both metrics together, and watching the spread between them, reveals whether retention is broad-based or propped up by a small number of customers.

What period should the NRR cohort cover? Match the cohort window to the company's contract structure. Monthly-billing SaaS typically uses a monthly cohort with month-over-month or trailing-twelve-month reporting; annual-contract enterprise SaaS uses an annual cohort. The cohort must be locked at the start date, and every movement tracked forward belongs only to those original accounts.

Sources

• Bessemer Venture Partners — State of the Cloud & SaaS Metrics
• OpenView Partners — SaaS Benchmarks Report
• KeyBanc Capital Markets — SaaS Survey
• SaaS Capital — Retention Benchmarks Research
• a16z — 16 Startup Metrics
• ChartMogul — SaaS Metrics Definitions
• Klipfolio — Net Revenue Retention Metric Guide
• Bain & Company — Net Revenue Retention and Durable Growth

Related on PULSE

• SaaS: Net Revenue Retention as the True North Star for Expansion MRR in 2027
• Net Revenue Retention (NRR) for SaaS: Churn Mitigation as a Growth Metric in 2027
• Top 10 Banking Net Interest Margin Revenue KPIs in 2027
• What are the key sales KPIs for the Commercial B2B SaaS industry in 2027?
• Top 10 SaaS Monthly Recurring Revenue Metrics to Track in 2027
• Top 10 Fintech Net Interest Margin and Revenue Efficiency Ratios in 2027