Gym KPIs and Benchmarks: The Numbers That Decide Survival

An independent gym lives or dies on about eight numbers: average revenue per member, length of engagement, retention, attendance, failed payments, coach pay, owner pay and how many visitors turn into members. Most owners can quote their member count and their rent. Far fewer can tell you their real monthly churn, and that is the number that decides whether next year is better or worse than this one.

Key takeaways

Key takeaways

1. Calculate length of engagement as 1 ÷ monthly churn. Software averages of current members almost always overstate it.
2. Two-Brain's 2025 data puts strong group-class gyms at US$187+ ARM and more than 25 months LEG.
3. Average annual retention across HFA's 2025 sample was 66.4%. A coached gym should aim to beat it.
4. Attendance falls before cancellations do. Run a weekly at-risk list from check-in data.
5. Track failed payments and recovery. Health and fitness firms in a 2023 survey reported losing 11% of revenue to them.
6. At CrossFit BC, 1.6% of 607 drop-ins in a year became paying members, so visitor follow-up has to be a system.

I co-own CrossFit BC in Vancouver, I was COO at Fitness Town while it grew from three locations to eight, and I have spent nearly 30 years in this industry. This page is the reference I wish someone had handed me early on. Every benchmark below comes from a named, dated source, and every formula is simple enough to run on a Sunday night with your gym software open.

One caution before the numbers. Most published benchmarks come from US gyms or from large club chains in many countries. Canadian independents have different rent, wages, taxes and payment systems. Use the benchmarks to find your blind spots, then judge yourself against your own trend line.

Which gym KPIs actually matter?

Here is the short list, with the best current benchmark I could verify for each.

A note on currency. The Two-Brain coverage does not state a currency, but its dataset comes from US software platforms and its LEG figures are explicitly from US gyms, so I read those dollar bands as US dollars. Convert before you compare.

Average revenue per member: what is a good number?

Average revenue per member (ARM) is your total monthly membership revenue divided by the number of paying members that month. PushPress, which builds a Two-Brain report into its software, defines it as membership payments (excluding free services) divided by paying members, with multi-month prepayments spread evenly across the months they cover.

In Two-Brain's 2025 State of the Industry data, reported by the Morning Chalk Up in November 2025, group-class gyms (CrossFit, bootcamp, class-based HIIT) sorted like this:

• Above average: US$187 or more a month, up from US$169 in 2024
• Average: US$126 to US$186
• Below average: US$125 or less

Small-group and semi-private gyms reached the top band at US$288 or more, and one-on-one personal training gyms at US$360 or more. That gap is the whole argument for adding a higher-touch tier: more coaching time per client lifts ARM faster than any price increase on a group membership.

What I watch: ARM by membership type, not one blended number. A gym with a healthy blended ARM can still be propped up by a handful of personal training clients while the core class membership quietly loses money. If your ARM falls while your member count rises, you are discounting your way to growth, and that rarely ends well.

Deeper reading: how to make gym membership pricing decisions, and if you are weighing a new programme to lift ARM, whether a HYROX affiliation makes sense for a small Canadian gym.

How do you calculate length of engagement honestly?

Length of engagement (LEG) is how many months a member stays, on average. Two-Brain's 2025 bands for group-class gyms, using Wodify data from US gyms:

• Above average: over 25 months
• Average: 15 to 24.9 months
• Below average: under 15 months

Chris Cooper, Two-Brain's founder, put it bluntly in that coverage: most gym owners overestimate their LEG.

The reason is the method. Most software takes every current member, adds up how many months each has been with you, and divides by the head count. That flatters you twice. Your ten-year veterans drag the average up, and everyone who already quit is missing from the sample. In a 2022 article Cooper showed how this method made gyms look far healthier than they were, and proposed a better one:

LEG = 1 ÷ monthly churn rate

Monthly churn = members lost this month ÷ paying members at the end of last month.

Use a rolling six-month average of churn so one bad January does not wreck the picture. A gym losing 4% of members a month has a LEG of 25 months. At 5% it is 20 months. At 7% it is about 14 months, which lands in the red band.

Two practical rules make the number honest. Count a member as lost when they stop paying for a membership, including the ones who "paused" and never came back. And count only paying members in the denominator, so trials and free passes do not dilute your churn.

Then multiply: lifetime value = ARM × LEG. That is the most you can rationally spend to win a new member, and the clearest way to see what a one-point drop in churn is worth. As an illustration only, at US$180 ARM, cutting monthly churn from 5% to 4% moves lifetime value from US$3,600 to US$4,500 per member.

What is a good gym retention rate?

The Health & Fitness Association's 2025 Fitness Industry Benchmarking Report put average annual member retention at 66.4%. The survey ran from April to June 2025 and covered 175 companies operating more than 17,000 facilities in 27 countries, with financial figures for 2024. In plain terms, about one member in three is gone within a year at the average operator.

That benchmark comes mostly from larger clubs and chains, many of them access-style gyms. A coached group-class gym should beat it, because members have relationships with coaches and each other. As a rough cross-check: 66.4% annual retention works out to around 3.4% monthly churn if losses were spread evenly through the year, which is a LEG of about 30 months. If your coached gym is under that, the problem is fixable and worth fixing first.

Retention also hides inside your new-member numbers. A gym that signs 15 people a month and loses 14 feels busy and goes nowhere. Track new members and lost members side by side, every month.

Deeper reading: gym retention benchmarks and a retention system for Vancouver gyms.

Why is attendance the leading indicator?

Retention and LEG tell you what already happened. Attendance tells you what is about to happen.

Les Mills analysed DataHub records from 601 facilities and published the results in 2019. Among live members, those who attended no Les Mills group classes had an average membership length of 23.4 months. Members attending once a month averaged 25.0 months, twice a week 31.9 months, and three or more times a week 35.3 months, about 51% longer. Members who attended three or more times a week also averaged 5.9 total visits a week.

A larger 2025 Les Mills report with ROR Partners and 4GLOBAL tracked 2.6 million member journeys across 1,312 clubs in the US and Europe from January 2024. Its most engaged members stayed 39% longer than typical gym members (23 months against 16) and were 88% more likely to still be active after 12 months.

Both are observational studies from a company that sells group fitness programming, so read them as associations, not proof that classes cause people to stay. Still, the pattern is consistent with what every coach sees on the floor: frequency comes first, and the cancellation email comes weeks later.

What to track: visits per member per week, and a list of every member whose attendance has dropped by half or more versus their own previous four weeks. That list is your retention to-do list for the week.

Deeper reading: what a great front desk does for member experience.

How much do failed payments cost a gym?

Failed payments are the leak most owners never measure. A member's card expires, the bank declines a charge, or a pre-authorized debit bounces. Nobody follows up for six weeks, and by then the member has drifted. That is involuntary churn: you lose someone who never decided to leave.

The best dated figures I could verify:

• A 2023 PYMNTS and FlexPay survey of more than 200 subscription executives found health and fitness companies lost 11% of top-line revenue to failed payments, against 9% across all subscription businesses. Only 17% of subscription businesses tracked failed payments at all. Those that did lost 37% less revenue and recovered 43% more of the failed payments.
• ABC Fitness, a gym software company, reported in December 2025 that across more than 40 million members at 30,000 gyms, studios and clubs, neobanks were a small share of bank debit transactions but over 26% of returns. Its year-end report put neobanks at 17% of new joins in 2025. Those are mostly US apps, so the specific risk differs in Canada, but the lesson travels: some payment methods fail far more often than others.
• ABC's own statistics page, updated July 2026, says failed payments drive up to one in three gym cancellations, citing its 2025 Wellness Watch. Treat that as an upper bound from a vendor that sells billing tools.

What to do: pull a report of every failed payment in the last 90 days and track three numbers. How many failed, how many you recovered, and how many of those members cancelled within 60 days. Then fix the process: card updater turned on, a retry schedule, a same-day personal message from a human, and a front desk that asks members to update expiring cards in person.

Deeper reading: the operational mistakes that quietly cost gyms members.

What should coach pay and staffing cost?

Staffing is usually your largest cost after rent, and in BC it is rising by law. The general minimum wage went to $18.25 an hour on June 1, 2026, up from $17.85, according to the BC government.

Benchmarks for the whole wage line:

• Innovation, Science and Economic Development Canada's industry data for small and medium fitness and recreational sports centres (revenue of $30,000 to $5 million, reference year 2024) shows labour and commissions at 26.1% of revenue, plus wages and benefits at 3.2%. Owners who pay coaches as contractors often book that cost as labour or commissions, which is why both lines matter.
• Statistics Canada reported that the whole fitness and recreational sports centre industry earned $5.8 billion in operating revenue in 2024 and spent $2.1 billion on salaries, wages, commissions and benefits. By my arithmetic, that is roughly 36% of revenue.
• Two-Brain's 2025 data put the average annual income of the highest-earning non-owner staff member at US$34,435 in group-class gyms, US$37,848 in small-group gyms and US$48,961 in one-on-one gyms.

That last figure is the uncomfortable one. The best-paid coach at an average group-class gym earns about US$34,000 a year, a hard number to live on in Vancouver. Coaches leave, and members often follow the coach out the door. Coach pay is a retention cost as much as a payroll cost.

The useful KPI is coaching cost per class and revenue per coaching hour. Divide your monthly membership revenue by the number of coached hours you delivered. If that number does not comfortably cover a fair hourly rate plus employer costs, you either have too many low-attendance classes or your ARM is too low.

A legal note: whether a coach is an employee or a contractor depends on the actual working relationship, not what the agreement calls it. This is general information; confirm your setup with an accountant or employment lawyer.

Deeper reading: paying coaches in BC.

How much should a gym owner pay themselves?

Two-Brain calls this "owner benefit": the money left for the owner after every expense, whether it is taken as salary or left in the business as profit. Its 2025 bands for group-class gyms:

• Above average: US$6,500 or more a month
• Average: US$2,800 to US$6,499
• Below average: US$2,799 or less

Small-group gyms reached the top band at US$7,000 a month and one-on-one gyms at US$8,500. The same data put above-average monthly revenue for group-class gyms at US$25,000 or more, with the bottom band under US$13,500.

On the Canadian side, ISED's 2024 figures for small and medium fitness centres show 75% were profitable, with an average net profit margin of 12%. Statistics Canada put the whole industry's operating margin at 8.3% in 2024, up from 5.8% in 2023. The HFA's larger operators reported a median EBITDA margin of 23.6%, which shows what scale and systems can do.

My rule: put the owner's pay on the profit and loss statement as a line item, at a figure you could hire a general manager for. If the gym only shows a profit because you work 60 hours for free, it is not profitable yet. It is subsidised by you.

Do drop-in and ClassPass visitors become members?

Rarely, on their own. The most cited study is old: the 2015 AFS studio benchmarking report, quoted by ClubIntel in 2016, found studios converted an average of 6% of unique ClassPass visitors into a membership or class pack, with some as low as 2%. ClassPass has since said, in 2025 coverage by Athletech News, that 94% of its users are new to the studios they visit, but it did not publish a conversion rate.

Here is our own funnel at CrossFit BC, from our Wodify records for the 365 days to 12 September 2026:

It took an average of 15.7 days from first drop-in to becoming a lead. Of the 513 who never converted, 261 (51%) had BC phone area codes, 126 had numbers elsewhere in North America, 52 were international and 74 had missing or unclear numbers. Every one of the 607 left an email and 596 left a phone number.

Two caveats. Wodify does not tag which drop-ins came through ClassPass, so "drop-in" is the closest proxy and includes walk-ins from other channels. And the local versus visitor split is inferred from area codes.

The lesson I took: half our non-converting drop-ins were locals, we had their contact details, and nobody was systematically following up. So in September 2026 we started a four-touch follow-up over 12 days, for BC-local drop-ins only. Day 1, a text or email asking how the class went. Day 3, a phone call that asks about their training rather than selling. Day 7, an email with the free trial offer. Day 12, a short close. It runs off a weekly list of about five names. It is too early to report results.

What to track: visitor to lead, lead to client, client to member still paying after 90 days. The last one is the only one that pays rent.

Deeper reading: is ClassPass worth it for gyms, and the CrossFit BC members case study.

Run your own numbers: formulas and a one-page scorecard

You need one export from your gym software and one from your accounting system. Set aside an hour on the first Monday of each month.

The formulas:

• ARM = membership revenue this month ÷ paying members this month
• Monthly churn = members lost this month ÷ paying members at the end of last month (use a six-month rolling average)
• LEG (months) = 1 ÷ monthly churn
• Lifetime value = ARM × LEG
• Annual retention = members active 12 months ago who are still active today ÷ members active 12 months ago
• Attendance rate = total check-ins this month ÷ paying members ÷ weeks in the month
• At-risk list = members whose last four weeks of visits are half or less of their previous four weeks
• Failed payment rate = failed charges ÷ attempted charges; recovery rate = recovered ÷ failed
• Labour share = all coach and staff pay, including contractors and employer costs ÷ total revenue
• Owner benefit = revenue − all expenses, before anything you pay yourself
• Visitor conversion = new paying members from visitors ÷ unique visitors in the period

Copy this table into a spreadsheet and fill in the "You" column monthly.

If you do only one thing this week, calculate monthly churn the honest way and compare it with what your software reports as average membership length. The gap between those two numbers tells you how much of your confidence is real.

More of my writing for gym and studio owners lives in the consulting articles library.

If you would rather have a second set of eyes on your numbers, bring your last six months of exports to a free 20-minute call and we will find the one number that is costing you the most. Book a gym numbers teardown.