Optimized Sales Optimized Marketing Target Accounts For CROs For CFOs For CMOs Blog News Glossary Compare Tools About Schedule a Demo
Retention & Growth

Net Revenue Retention Denominator

ORM Technologies
Home/ Glossary/ Net Revenue Retention Denominator
Definition The NRR denominator is the recurring revenue of the customer cohort measured on the first day of the period. It stays fixed while the numerator moves, which is what makes the ratio a retention measure rather than a growth measure.

The NRR denominator is the recurring revenue of a fixed customer cohort on the first day of the measurement period. Everything the cohort does after that moment lands in the numerator. Locking the base is what separates a retention ratio from a growth ratio, and most disputes about NRR turn out to be disputes about this one number.

The base is a population and a date

Two decisions define it. Which customers were active on day one, and what each one was worth on that day.

Active means holding a live contract at the start of the period. Worth means contracted recurring revenue, annualized or monthly, applied the same way to every account in the cohort. Once those are set, the denominator never moves again for that period, regardless of what happens to the accounts inside it.

Four ways the denominator gets broken

Mid-period additions. A customer signed in week three gets swept into the base because the query filtered on customers active at any point in the period. The base inflates, NRR falls, and nobody can explain why. One-time revenue. Implementation fees, training, and overage true-ups sit in the same billing table as subscriptions. Pull them into the base and the denominator carries revenue that was never going to repeat, which reads as contraction in the following period. Churn removal. Someone excludes accounts that cancelled during the period so the comparison is apples to apples. The result cannot go below 100%, and gross churn disappears from the picture entirely. Silent restatements. A backdated contract amendment or a data migration changes historical ARR, so last quarter's denominator no longer matches what was reported. Snapshot the base at close and store it rather than recomputing it on demand.

Reconcile the base against the waterfall

The base should tie to the beginning ARR line of your monthly ARR reconciliation. ORM runs that reconciliation month by month, where beginning ARR equals prior month ending ARR, then walks through churned customer ARR, churned product ARR, and product decrease ARR on the contraction side and new product ARR and increased product ARR on the expansion side. Gross and net revenue retention sit on the same chart.

When NRR is computed from a separate query rather than from that reconciliation, the two eventually disagree. The waterfall is the control, and any NRR figure that cannot be tied back to it should not reach a board deck.

Fix the base before benchmarking

Comparing your NRR against a published figure is meaningless until you know both sides used the same base rules. Write down the cohort date, the revenue types included, the currency treatment, and the handling of reactivated customers, then apply that definition to at least eight prior periods so the trend is built on one method.

A base that changes definition quarter to quarter produces a retention series nobody trusts, and untrusted inputs degrade forecast accuracy the same way stale pipeline does. Get the denominator right and net revenue retention becomes a number you can plan against.

Frequently Asked Questions

What exactly goes into the NRR denominator?

Contracted recurring revenue for every customer active on the first day of the period, measured on an annualized or monthly basis consistently across the whole cohort. One-time fees, professional services, and pass-through costs stay out, because none of them recur and including them makes the base drift for reasons that have nothing to do with retention.

Should the denominator change if a customer churns mid-period?

No. That is the entire mechanism. The base is locked on day one and the churn shows up as a reduction in the numerator. Removing churned accounts from the denominator produces a number that cannot fall below 100% and is worthless.

Do you use beginning ARR or average ARR as the base?

Beginning ARR. Averaging the base across the period blends the effects you are trying to isolate and makes the result impossible to reconcile against an ARR waterfall. Averages belong in churn rate variants where the population genuinely changes, not in a fixed-cohort retention ratio.

How do currency and price book changes affect the base?

Both can move NRR without any customer behavior changing. Hold the exchange rate constant across the period for multi-currency books, or restate the base at the current rate, and record the choice. Otherwise a currency swing reads as expansion or contraction that nobody sold and nobody lost.

Put these metrics to work

ORM builds custom revenue forecast models that turn concepts like net revenue retention denominator into prescriptive action for your team.

Schedule a Demo