How to calculate member retention rate (formula)

Retention rate has the ring of a statistics exam question, all cohorts and percentages, but the formula itself is about ten seconds of primary-school arithmetic. The awkward bit isn't the sum. It's the handful of quiet decisions around it - which members count, over what stretch of time, and what on earth to do with the people who joined halfway through - and getting those wrong is how a club ends up with a number that flatters it rather than tells it anything. This is the narrow, number-minded piece. If you want the tactics that actually shift the figure, that's the companion guide on increasing your retention rate, and the bigger question of why members stay in the first place is the complete retention guide. Here I just want to get you to a figure you can trust.

The formula, plainly

Here it is, and it really is this simple:

Retention rate = (members you kept ÷ members you started with) × 100

The only word doing any real work there is "kept". It means: of the members you had at the start of a period, how many were still with you at the end. That's it. The trap - and it's the one nearly everyone falls into the first time - is quietly letting new joiners sneak into the "kept" pile, which makes your rate look far healthier than it is. So before we go near the wrinkles, let me run a real set of numbers through it, because it lands better as arithmetic than as a definition.

A worked example - Napas Biru

Say you help run Napas Biru, a freediving and scuba community up in Amed, on the east coast of Bali - 240 members at the start of the year, a mix of weekend breath-hold divers and scuba folk who drive up to the Liberty wreck at Tulamben, with dues that keep the safety kit and the boat fuel covered, monthly pool-and-breath sessions and the odd shore clean-up. You want to know how well you held onto people over the year.

You start with 240. At the end of the year there are 250 names on the books. The tempting sum is 250 ÷ 240 = 104%, which would mean you kept more members than you started with, which is impossible, and that impossibility is the tell. You haven't calculated retention at all - you've calculated headcount, and headcount is hiding something. Because during the year, 70 brand-new people signed up. So of those 250 at year end, 70 were never part of your starting group. Strip them out and the original crowd still with you is 250 - 70 = 180. That's 180 of the 240 you began with, which gives you:

180 ÷ 240 × 100 = 75%

So your retention rate is 75%, not 104%. The other 60 - a full quarter of where you started - drifted off over the year, and the 70 new sign-ups very nearly papered over the hole. Written as a formula so you can reuse it, the version that keeps you honest is:

Retention rate = ((members at end − new members gained) ÷ members at start) × 100

or, in the shorthand you'll see about, (E − N) ÷ S × 100, where E is the count at the end, N is everyone who joined during the period, and S is where you started. For Napas Biru that's (250 - 70) ÷ 240, which is your 75%.

The wrinkles that trip people up

The sum is easy. These are the bits that actually decide whether your number means anything, and none of them is hard once someone points them out.

People who joined mid-period

We've met this one already - it's the reason the naive 104% was nonsense. A member who joined in March and is still there in December is a genuinely good outcome, but they're an acquisition, not a retention, and dropping them into a sum measured against a starting base that never included them just inflates the figure. Keep new joiners out of the numerator. If what you actually want to know is how well you hold onto new members specifically - which is well worth knowing, since the first months are where most leavers are lost - then measure them as their own cohort: take everyone who joined in, say, the first quarter, and check how many are still around a year on. That's a different and very useful number. Just keep it separate from the whole-club rate rather than blending the two.

Gross versus net

The 75% above is gross member retention - bodies kept, counted as heads. If your dues are the same for everyone, that's usually all you need. But if you run tiers (a full-dive membership, a cheaper social or snorkel tier), there's a money version worth a glance, because a member who drops from the full tier to the cheap one still counts as retained - they're still a member - even though you've lost a chunk of their dues. If the rupiah matters more to you than the raw count, run the very same sum on dues income rather than on people, and you'll sometimes find the two disagree: your member retention looks fine while your revenue retention is quietly sliding, or occasionally the other way about. For most clubs the head count is plenty. Just know the money version exists for the day the money question comes up.

What period to measure over

Match the window to your renewal cycle. Annual dues, measure over a year - it's the only period where the question "did they come back?" actually means something. Measure an annual membership month by month and you'll get a giddy 99% every single month, because almost nobody's renewal happens to fall due in any given four weeks, and you'll learn precisely nothing from it. Pick either the calendar year or a rolling twelve months, then use the same window every time. The comparability matters more than which one you choose - a 75% that was worked out the same way three years running tells you far more than a technically-perfect figure you calculate differently each time.

When is a member actually "lost"?

This is the genuinely messy edge, the human one. A member who's three weeks past due but always pays eventually - lapsed, or not? Someone who said they'd renew "next week" a month ago? Honestly, it barely matters which way you rule, as long as you rule once and apply it the same way every year. A common, sensible cut is that 30 - 60 days past due counts as gone. And the arithmetic doesn't care why someone left: the diver who moved to Jakarta for work, the one who took up golf, and the one who simply forgot all count identically as "not retained" (though it's worth glancing at the reasons separately, because some of them you can fix and some of them you never could).

Churn is the same number wearing a different hat

You'll see churn rate talked about as though it's a separate metric to learn, and it isn't - it's just retention told backwards. On the same group of members, churn rate = 100% − retention rate. Napas Biru's 25% churn and its 75% retention are one fact said two ways: 180 kept plus 60 lost is exactly the 240 you started with. Track whichever one you find easier to say out loud in a committee meeting. Some people are more moved by "we lost a quarter of them last year" than by "we kept three-quarters", and if the blunter framing gets the committee to actually do something about onboarding, use it. It's the same number either way.

So what counts as a "good" rate?

It depends, and I mean that honestly rather than as a dodge. There's no universal benchmark worth trusting, and anyone quoting you one for your particular kind of club probably hasn't run enough of them. A professional association that people need for their livelihood can sit comfortably in the high 80s or low 90s, because letting the membership lapse costs them something real. A casual community like Napas Biru will naturally run lower - people's hobbies genuinely change, and a keen freediver who relocates inland isn't a failure of your onboarding, they're just someone whose life moved. As a very loose feel: a lot of hobby and community memberships live somewhere in the 60 - 80% range, professional and career-linked ones higher, but treat even those as weather rather than law. The one comparison that's always fair is you against your own past self. A club going 72% to 75% to 79% over three years is winning, even though not one of those figures is a trophy - the trend is the honest scorecard, and the benchmark you read on someone's blog usually isn't.

Once you've got the number, the obvious next question is how to nudge it upward, and that's a whole piece of its own - I've set out which handful of things actually move the rate, and roughly how much each is good for, in the guide to increasing your retention rate. But you can't shift a number you haven't measured, so this is the place to start.

And start rough. Work the figure out once, even on the back of an envelope - members at start, members at end, knock off the new joiners, divide - write it on a page, and do the same thing next year in the same spot right underneath it. That one small habit, a single honest number a year sitting next to last year's, will tell you more about the health of your membership than any amount of clever tracking you never quite get round to setting up….

Working this out by hand once a year is no great burden, and plenty of clubs do exactly that with a spreadsheet and a bit of discipline. It's one of the things Anggota does quietly in the background - the starting count, the new joiners, the leavers and the retention rate sitting there year on year so you can watch it move without doing the sum yourself. If that sounds handy you're welcome to try it for free and see how it fits. And if you'd rather keep your numbers in whatever you already use, that's genuinely fine - the formula doesn't care what tool it lives in, and a retention rate you actually calculate and look at each year is worth having however you get to it.