Analytics
Free net and gross revenue retention calculator
By Charles Summers · Updated · Free, no signup
Short answer
Gross revenue retention is starting MRR minus churn minus contraction, over starting MRR, and it cannot exceed 100 percent. Net revenue retention adds expansion back and can. Neither includes revenue from new customers, because those customers were not in the base you are measuring retention on, and including them is the single most common way these numbers get inflated. This calculator shows both figures, the inflated version for contrast, and the gap between gross and net that tells you what expansion is hiding.
Use the net revenue retention calculator
What does this tool actually do?
Gross revenue retention is starting MRR minus churn minus contraction, over starting MRR, and it cannot exceed 100 percent. Net revenue retention adds expansion back and can.
It runs entirely in your browser. Nothing you type is sent to a server, no account is required, and there is no usage limit, because there is no cost per run to control.
Five movements, and only four of them belong in retention
Every month your recurring revenue moves in exactly five ways. Customers cancel, which is churn. Customers stay but pay less, which is contraction: fewer seats, a lower plan, less usage, a discount negotiated at renewal. Customers stay and pay more, which is expansion. Customers who left come back, which is reactivation. And customers you did not have before start paying, which is new. Ending revenue is the start plus all five, and that identity is the only piece of this that is not up for debate.
Retention asks a narrower question: of the revenue you had at the start of the period, how much do you still have at the end? That question is about a fixed group of customers, the ones already on the books when the clock started. New customers were not in that group. They cannot demonstrate that you kept anything, because there was nothing yet to keep. Add their revenue to the numerator and you have stopped measuring retention and started measuring growth wearing retention's label.
The size of the distortion is easy to feel with real figures. Start at 500,000 dollars of MRR, lose 25,000 to cancellations and 10,000 to downgrades, gain 30,000 from expansion and 60,000 from new customers. Correctly computed, net revenue retention is 495,000 over 500,000, or 99 percent, and the honest reading is that the existing base is shrinking slightly. Fold the new revenue in and you report 111 percent, which describes a completely different company. The 12-point gap is entirely the sales team's new business, which nobody was arguing about.
What makes this error so persistent is that the inflated number cannot easily go wrong. While you are growing at all, new revenue exceeds losses, so the metric sits above 100 percent permanently and never delivers the warning it exists to deliver. A retention figure that only turns bad when the company is already visibly failing is not a leading indicator, it is a decoration. Lock the cohort at the start of the period, track only what that cohort did, and let new business be reported as new business.
The gap between gross and net is a diagnosis, not a rounding difference
Gross revenue retention counts losses only, so it is capped at 100 percent and is best read as your leak rate. Net revenue retention adds expansion, so it can exceed 100 and is best read as whether the existing base grows itself. Subtract one from the other and you have your expansion rate expressed in points, which is the number that actually tells you what kind of business you are running.
Consider two companies both reporting 112 percent net retention. The first has gross retention of 96, so it barely leaks and expansion is a genuine bonus. The second has gross retention of 78 and a 34-point expansion engine papering over it. On the headline metric they are identical. In practice the second one is renting its growth from a handful of accounts that happened to expand, and expansion is far more concentrated than churn: in most software businesses the top decile of accounts produces the large majority of upsell, so a single account changing its mind moves the whole number.
The two also behave very differently when conditions tighten. Expansion is the first thing to disappear in a downturn, because seat growth stops when hiring stops and usage-based revenue falls when customer volumes fall. Net retention therefore collapses towards gross retention exactly when you need the cushion, which is why gross retention is the more defensive of the two numbers and the one worth reporting first. A business planning on 115 percent net retention with 80 percent gross retention has planned for good weather.
There is also an ordering trap in how expansion gets counted. If a customer doubles their seats in March and cancels in June, the expansion sits in one month and the whole enlarged amount leaves in another, so a quarter measured carelessly can show both a strong expansion figure and an intact retention figure while the account is already gone. Recording each movement in the month it occurs, against the cohort that was locked at the start, is what keeps the arithmetic honest across periods.
Contraction is the line most teams never separate out
Contraction is the quiet one. Nobody sends a cancellation email when they drop from 40 seats to 28, and no alert fires when a usage-based account halves its volume. The revenue leaves anyway. Teams that do not track contraction as its own line typically do one of two things with it, and both destroy information. Some fold it into churn, which makes churn look worse than it is and points the investigation at cancellations that did not happen. Others net it against expansion and report the difference, which makes the loss vanish entirely inside a positive number.
Separating it out changes what you conclude. High churn with low contraction says people are deciding your product is not worth having. Low churn with high contraction says they still want it but are buying less of it, which is usually a pricing, packaging or customer-health problem rather than a product one, and the fixes are entirely different. In seat-based products contraction tracks your customers' headcount more than their opinion of you, so it moves with their hiring plans and with the economy, and treating that movement as a product failure sends the team to fix the wrong thing.
Double counting is the other hazard. A customer who downgrades in one month and cancels in the next should show the downgrade as contraction when it happens and only the remaining amount as churn when they leave. Recording the original full amount as churn on the way out counts the same dollars twice and quietly overstates your total losses. If your movement lines do not reconcile back to the change in total recurring revenue, this is almost always where the discrepancy is hiding.
One practical habit is worth the trouble: reconcile every month. Starting revenue plus new plus expansion plus reactivation minus churn minus contraction should equal ending revenue exactly. If it does not, the retention figures built on those movements are not wrong by a little, they are unaudited, and every conclusion drawn from them inherits the gap.
Periods, denominators and why quoted benchmarks rarely compare
The most common comparability failure is period mismatch. A monthly net retention of 101 percent is not the same claim as an annual one: compounded over twelve months it is about 112.7 percent. A monthly 98 percent compounds to roughly 78.5 percent annually. Public company benchmarks are almost always annual, and internal dashboards are almost always monthly, so a team can spend a quarter feeling comfortable against a benchmark they are not actually being measured against. Convert by raising to the twelfth power, never by multiplying by twelve, which is meaningless for a rate that compounds.
The denominator is the second choice that changes the answer. Retention measured against the start of a rolling twelve-month window behaves differently from retention measured on a signup cohort followed forward, and cohort figures are the more informative of the two because they separate the behaviour of last year's customers from this year's. Snapshot figures blend everyone together, which flatters a business whose older customers are its stickiest, and that is most businesses.
Then there are the definitional liberties in reported numbers. Some published retention figures are computed only across customers who were still customers at the end of the period, which mathematically removes churn from the calculation and produces a number that can barely go below 100. Others exclude accounts below a revenue threshold, or measure at the parent-company level so that one division leaving does not register. None of these are necessarily dishonest, but they are not comparable to a figure computed over the whole book, and the definition is rarely printed next to the number.
Because of all that, treat published ranges as loose context rather than a target. Median net revenue retention across public software companies has generally been reported somewhere near or a little above 100 percent, with strong performers well above and the whole distribution shifting downwards after 2022 as seat counts and usage contracted. Any source quoting a single tidy figure to the decimal point is describing a specific sample with a specific definition. Your own trend, measured the same way every month, is worth more than any of them.
Numbers worth knowing
| Metric | Typical | What it means |
|---|---|---|
| Gross revenue retention ceiling | 100%, always | It counts only losses, so it cannot exceed the base. Any GRR above 100 means expansion has been let into the numerator by mistake, which is worth checking before anything else. |
| Net revenue retention baseline | 100% holds flat | At exactly 100 the existing base replaces its own losses through expansion. Below it, new business is buying replacement before it buys growth. |
| Published NRR ranges | near or a little above 100% median, wide spread | Reported medians moved down after 2022 as seat counts and usage fell. Definitions vary enough between filings that a precise cross-company figure is not worth quoting. |
| Monthly to annual conversion | raise to the 12th power | Monthly 101% is about 112.7% a year; monthly 98% is about 78.5%. Multiplying a compounding rate by twelve produces a number that means nothing. |
| SaaS quick ratio | 4 is the usual growth-stage marker | New plus expansion divided by churn plus contraction. A rule of thumb from growth-stage investing, useful because it fails loudly when losses start outrunning gains. |
Mistakes that quietly cost you results
- Including new-customer MRR in the retention numerator
- Those customers were not in the base you are measuring, so their revenue proves nothing about retention. Including it also means the metric stays above 100 percent for as long as you are growing at all, which removes the only warning it exists to give.
- Reporting net retention without gross retention beside it
- Net alone cannot distinguish a business that barely leaks from one with a 34-point expansion engine covering a severe leak. Publish both, because the gap between them is the expansion rate and it is the first thing to disappear in a downturn.
- Folding contraction into churn, or netting it against expansion
- Folding it in blames cancellations that never happened; netting it out makes the loss invisible. Keep it as its own line, because high contraction with low churn is a pricing and packaging problem, not a product one.
- Counting a customer's full original MRR as churn after they had already downgraded
- That counts the same dollars twice. Record the downgrade as contraction in the month it happened and only the remaining amount as churn on the way out, then reconcile the movements back to the actual change in total MRR.
- Comparing a monthly retention figure to an annual benchmark
- Compound it first. Monthly 98 percent is annual 78.5 percent, which is a different conversation entirely, and multiplying by twelve to annualise a compounding rate is not an approximation, it is arithmetic that does not apply.
What does the output look like?
This is the exact output the tool produces from the example inputs. It is generated by the same code that runs when you click the button, so what you see here is what you get.
Frequently asked questions
Why is new-customer MRR excluded from both retention numbers?
Because retention measures what happened to the revenue you already had. A customer who signed up during the period was not in the starting base, so their money cannot demonstrate that you retained anything. Including it also breaks the metric as a warning system: while the company is growing at all, new revenue exceeds losses, so the figure stays above 100 percent permanently and only turns bad once the business is visibly failing.
What does the gap between gross and net retention tell me?
It is your expansion rate in points, and it is the most diagnostic number on the page. Two companies at 112 percent net retention are entirely different businesses if one has gross retention of 96 and the other 78: the second is covering a serious leak with upsell that is usually concentrated in a handful of large accounts. Expansion also disappears first when customers freeze hiring, so a wide gap is a fair-weather structure.
Can gross revenue retention ever be above 100 percent?
No. It counts only cancellations and downgrades against the starting base, so the numerator is always smaller than the denominator. If your calculation returns something above 100, expansion revenue has been included by mistake, or a reactivated customer has been counted as retained rather than as a returning logo. It is the fastest sanity check available on a retention model.
Are these figures monthly or annual?
They are whatever period your inputs cover, which is usually a month. To compare against published benchmarks, which are almost always annual, raise the monthly figure to the twelfth power rather than multiplying it. Monthly 101 percent becomes about 112.7 percent a year and monthly 98 percent becomes about 78.5 percent, so the two framings can lead to opposite conclusions about the same business.
Where does the quick ratio in the output come from?
It is new plus expansion divided by churn plus contraction, a growth-stage investing rule of thumb that asks how many dollars of gain you generate per dollar lost. Four is the commonly cited marker for a healthy fast-growing company. Its value is that it is blunt: as soon as losses start outrunning gains it drops below one, regardless of how the retention percentages have been defined.
Related free tools
- LTV to CAC Ratio Calculator Analytics
- Pipeline Coverage Calculator Analytics
- Marketing Budget Allocator Analytics
- Lead Scoring Model Builder Sales
- Cold Email Generator Outbound
Some links on this site are affiliate links, which means Hacking Demand may earn a commission if you buy through them at no extra cost to you. This does not influence which tools are listed. The tools on this page are free and have no affiliate relationship of any kind.