Key Takeaways
- The median private B2B SaaS company retains 91% of its recurring revenue on a gross basis and 101% on a net basis, according to SaaS Capital’s 2025 survey. Companies with contracts above $250,000 a year run higher, with median gross retention of 95%.
- Gross revenue retention (GRR) shows how much of your starting revenue you kept, with no credit for upsells. Net revenue retention (NRR) adds expansion back in. Investors and buyers read GRR first, because a strong NRR can sit on top of a customer base that is leaking.
- A “good” number is one at or above your direct competitors, calculated the way a diligence team would calculate it .Most companies we work with don’t track retention figures at all before our CFOs start in.
You are two weeks from a board meeting and the draft deck says net revenue retention is 108%. An investor on the pre-call asks a different question: what is gross retention, and how has it moved over the last four quarters? Nobody has the number. Your head of customer success thinks it is in the low nineties. Your controller isn’t sure which customers count.
That exchange happens in most SaaS companies between $3M and $20M in annual recurring revenue (ARR), and it matters more than it seems. Retention is the number investors and acquirers use to judge whether your revenue is durable. If you can state it precisely, show how it is built, and say where you stand against your market, you walk into a raise or a sale process with an advantage. This article covers what the two metrics measure, what the benchmarks are, how the calculation commonly goes wrong, and what to do about it.
What do GRR and NRR measure?
Both metrics start with the same group: the customers you had at the beginning of a period, usually 12 months ago, and the recurring revenue they were paying then.
Gross revenue retention is the share of that starting revenue you still have from those same customers, after subtracting cancellations (churn) and downgrades (contraction). It ignores upsells entirely, so it can never exceed 100%.
Net revenue retention starts from the same place and then adds expansion: upsells, added seats, price increases, and cross-sells to those same customers. Because of that, it can exceed 100%.
Neither metric includes revenue from customers you signed during the period. That is the most common mistake, and we will come back to it.
Here is a worked example. On January 1 your customers represent $5,000,000 of ARR. Over the next 12 months, customers worth $350,000 cancel, others downgrade by $100,000, and others expand by $600,000. You also sign $1,200,000 of new customers.
- GRR is ($5,000,000 less $350,000 less $100,000) divided by $5,000,000, which is 91%.
- NRR is ($4,550,000 plus $600,000) divided by $5,000,000, which is 103%.
- The $1,200,000 of new business belongs in your growth rate and appears in neither retention figure.
What is a good retention rate?
SaaS Capital surveys private B2B software companies every year, and its 2025 retention benchmarks put the median at 101% net and 91% gross. Contract size changes the picture. Companies with annual contract values above $250,000 report median gross retention of 95%, while those below that level sit around 91%. The same research found that growth rate rises sharply with net retention, and it describes gross retention as table stakes: you need it to be solid before anything else counts.
Those figures are for private companies, which is the right comparison for most founders. Public SaaS benchmarks run higher and reflect much larger customers.
Benchmarks give you a range. The more useful answer comes from your own market. Mark Kearney, one of our fractional CFOs who works mostly with software companies, hears this question from SaaS founders all the time:
“When we say things like gross revenue retention at 96%, is that a good number or a bad number? It really comes down to: in your marketplace, what are your competitors doing?”
If your competitors hold 98%, then 96% is a problem to fix. If they are at 87%, it is a strength to lead with. Buyers and investors are almost always choosing between you and a comparable company, and they compare on a shared set of metrics. A founder who knows where the company stands against peers controls that conversation.
Why investors read gross retention first
Net retention gets the attention because it connects directly to valuation. McKinsey’s analysis of more than 100 B2B SaaS companies found that those in the top quartile of valuation multiples had NRR of 113%, against 98% for the bottom quartile, and traded at a median of 24 times revenue compared with 5 times. A separate McKinsey study of public B2B SaaS companies found a median multiple of 21 times revenue for those with NRR of 120% or more, against 9 times for the rest.
Experienced investors still ask for GRR first, for a simple reason. Two companies can both report 105% NRR. One keeps 95% of its base and expands a little. The other loses 20% of its base every year and makes it up by selling hard into the customers who stay. The second company has to replace a fifth of its revenue annually before it grows at all, and its expansion depends on a shrinking pool. The net figure is identical, and the businesses are very different.
Reporting both, side by side and over time, tells the full story. A company that shows only NRR invites the question of what GRR would reveal.
Five ways the calculation goes wrong
When our CFOs take on a new SaaS client, the retention figure in the existing board deck often changes once it is rebuilt from the billing data. These are the errors behind most of the differences.
1. New customers are in the numerator
Comparing total ARR today with total ARR a year ago measures growth. Retention only counts customers who were there at the start. Mixing in new logos inflates the result, often by a wide margin.
2. Logo retention is reported as revenue retention
Keeping 92% of your customers and keeping 92% of your revenue are different facts. If the customers who leave are your smallest, revenue retention will be higher than logo retention. If one large account leaves, it will be much lower. Diligence teams ask for both, and each should carry its own label.
3. ARR doesn’t tie to anything
ARR is an operating metric, and it has no official accounting definition. That makes it easy for a spreadsheet version of ARR to drift away from what customers are billed. Retention built on that spreadsheet inherits the drift. Your ARR schedule should reconcile to invoicing (often using available tools like Maxio or Chargebee to help keep track) and, with explainable differences, to recognized revenue under GAAP. We cover the accounting side in deferred revenue for SaaS.
4. Churn is recorded at the wrong time
A customer who gives notice in March on a contract ending in December is churned revenue you already know about. Some companies record it in March, some in December, and some whenever the last invoice is paid. Pick a rule, write it down, and apply it every period. Multi-year contracts need the same care: a customer in the second year of a three-year deal had no decision to make this year, so including them makes retention look better than your renewal experience. Many investors will ask for retention measured only on contracts that came up for renewal.
5. One blended number hides the segments
A company-wide GRR of 90% can be 96% among mid-market customers and 78% among small ones. That is two different businesses, and the right response to each is different. Break retention out by customer size, by product, and by the year the customer signed, so you can see where revenue is being lost.
Speak to a CFO
If a raise, a board meeting, or a sale process is coming and you aren’t certain your retention numbers would survive diligence, that is worth an hour now. Our CFOs rebuild SaaS metrics from the billing data and tell you where you stand. Book a CFO strategy call with Ascent CFO Solutions.
How to build retention reporting you can defend
The tool for this is an ARR waterfall, updated every month. It starts with opening ARR and shows each movement on its own line: new customers, expansion, contraction, churn, and closing ARR. Every line traces to specific customers and specific invoices. GRR and NRR fall straight out of it, and so does the answer to any question a board member asks.
Four practices make the waterfall trustworthy:
- Write a one-page definitions memo covering what counts as recurring, when churn is recorded, and how discounts, pilots, and usage fees are treated. Diligence teams ask for this, and having it ready signals a finance function that knows its numbers.
- Build it from the billing system, and reconcile it to the general ledger each month as part of the close.
- Report on a trailing 12-month basis every month, so the board sees a trend line.
- Carry the retention assumption into the forecast. A 24-month model that assumes 95% gross retention when history shows 89% overstates revenue in every future period.
Once the data is clean, it belongs on a dashboard next to the other metrics investors compare companies on: net new ARR, customer acquisition cost payback, the SaaS magic number (new recurring revenue generated per dollar of sales and marketing spend), and burn. Our articles on SaaS metrics that increase your company’s value and the metrics that matter at seed and Series A cover the rest of that list, and building real-time dashboards from your source systems covers the mechanics.
When a retention problem is a finance problem
Most churn is about the product and the customer relationship. A meaningful share is financial, and a CFO can find it. Failed credit card payments that nobody follows up on show up as churn. So do renewals that lapse because the contract end date lived in a salesperson’s inbox. Discounts granted at renewal with no approval process show up as contraction. Price increases that were written into contracts and never invoiced show up as expansion you didn’t collect.
Each of these is fixed with a renewal calendar, a billing process someone owns, and a monthly review of every line on the waterfall. For a company at $5M of ARR, recovering even one point of gross retention is $50,000 of recurring revenue every year, and it compounds.
Get right-sized financial leadership from experienced CFOs ready to lead your team.
Frequently asked questions
What is a good gross revenue retention rate for SaaS?
For private B2B SaaS, the median is about 91%, and companies selling large contracts (above $250,000 a year) have a median around 95%, per SaaS Capital’s 2025 benchmarks. Above 90% is generally considered healthy for small and mid-market customers, and enterprise-focused companies are expected to be in the mid-nineties. Compare yourself with companies that have similar contract sizes and customers.
What is the difference between gross retention and net retention?
Gross retention measures how much starting recurring revenue you kept after churn and downgrades, and it tops out at 100%. Net retention takes the same starting revenue and adds expansion from those customers, so it can exceed 100%. The gap between the two is your expansion revenue.
Can gross revenue retention be over 100%?
No. Because GRR excludes all expansion, the best possible result is keeping every dollar you started with, which is 100%. If a report shows GRR above 100%, expansion or new customer revenue has been included by mistake.
Do investors care more about NRR or GRR?
They look at both, and many read GRR first. NRR drives growth and valuation. GRR shows whether the underlying customer base is stable, and a weak GRR raises the question of how long expansion can keep covering the losses.
How often should a SaaS company report retention?
Calculate it monthly on a trailing 12-month basis and include it in every board package. Monthly snapshots let you see a trend forming two or three quarters before it would show up in an annual figure.
How do you calculate retention with usage-based pricing?
Usage-based companies usually measure the trailing 12 months of revenue from a customer group against the prior 12 months from the same group, since there is no fixed contract value to compare. Usage naturally rises and falls, so document the method and apply it consistently, and expect investors to ask how much of the movement is seasonal.
Know the number, and know how it was built
We help founders and CEOs of SaaS and technology companies across the country get their metrics investor-grade before a raise, a board meeting, or a sale. Our fractional CFOs build the ARR waterfall, the forecast, and the board reporting, and our data team automates the dashboards. You can read more about our work with SaaS and technology companies, how we support capital raises, and what to look for in a fractional CFO for a SaaS company.
Contact Us
Questions or business inquiries regarding our part-time CFO, finance and accounting services are welcome at: info@ascentcfo.com
