Monthly NRR compares a cohort's revenue at the start of a month against its revenue at the end of that month. Annual NRR does the same across twelve months. Both are correct. They measure different windows, and the relationship between them is not arithmetic.
Why the two numbers do not reconcile
Monthly NRR re-bases every month. Each new period starts a new cohort at the prior month's ending value, so the effect of a customer that expanded in January is baked into February's denominator rather than compounding on top of it.
Annual NRR holds one cohort for the full twelve months. Every renewal, expansion, downgrade, and cancellation within that window lands against the same fixed base. In an annual contract business, most accounts have exactly one contract event per year, and the annual window is the only window that reliably contains it.
| Monthly NRR | Annual NRR | |
|---|---|---|
| Cohort | Re-based each month | Fixed for twelve months |
| Captures renewals | Only in the month they occur | All of them |
| Volatility | High, driven by renewal date clustering | Low |
| Best use | Operating review, early warning | Board reporting, valuation conversations |
Renewal date concentration drives the gap
If a large share of your contracts renew in one or two months, monthly NRR will spike and crater around those months while the rest of the year reads flat. That pattern says nothing about customer health. It says your renewal calendar is concentrated.
Before reacting to a monthly reading, check what proportion of the cohort had a contract event that month. A month where almost nothing was up for renewal produces a number with no information in it, and treating it as a trend point is how teams talk themselves into problems that do not exist.
Run the monthly waterfall, report the annual ratio
The practical setup is to reconcile ARR monthly and derive both figures from it. ORM builds this as a month by month walk where beginning ARR equals the prior month ending ARR, then moves 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, with gross and net retention read off the same chart.
That reconciliation gives you the monthly detail for operations and a clean trailing twelve month roll-up for the board. Both numbers come from one source, so nobody spends the meeting arguing about which query was right.
Match the window to the decision
Use monthly NRR to catch a contraction pattern early enough to act inside the quarter. Use annual NRR to judge whether the installed base compounds, to compare against outside figures, and to feed the retention assumption in your revenue model.
The failure mode is reporting the monthly number as if it were the annual one. It reads lower or higher depending on the calendar, it moves for structural reasons, and once a leadership team starts steering on it, net revenue retention stops informing the plan and starts distorting it. For the mechanics of turning a retention assumption into a forward number, see how to forecast revenue.
Frequently Asked Questions
Can you annualize monthly NRR by compounding it?
No. Compounding twelve monthly readings assumes each month starts from the previous month's cohort, but the annual figure holds a single cohort fixed for twelve months. Renewals cluster on anniversary dates, so the monthly series misses most contract events while the annual window captures all of them.
Which NRR should go in the board deck?
Trailing twelve month NRR, because it captures a full renewal cycle for every account in the cohort. Monthly NRR belongs in the operating review, where it functions as an early warning rather than a headline. Reporting the monthly number to a board invites questions about volatility that the metric was never built to answer.
Why is monthly NRR so volatile in an annual contract business?
Because most months contain no contract events for most accounts. Revenue changes concentrate on renewal dates, so a month with a heavy renewal cohort swings hard while quiet months read near 100%. The volatility is calendar structure rather than customer behavior.
Does monthly NRR make sense for annual contracts at all?
Yes, as an operational signal. Month by month it shows expansion and contraction landing as they happen, so a bad renewal cohort is visible within weeks instead of at the end of a twelve-month window. Read it as a leading indicator and report the annual figure as the result.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like monthly vs annual net revenue retention into prescriptive action for your team.
Schedule a Demo