Gym Marketing: Count the Members Who Stay

Gym marketing works when the members it brings are still paying dues months later, not when someone signs a waiver. Count members still active at 3, 6 and 12 months by the channel that brought them, and a referral worth repeating separates itself from an ad worth cutting.

Quick answer

Judge every channel, trial and promo by members still paying at 3, 6 and 12 months, never by joins alone. Pull your membership software's join and cancellation report by source for a cohort that is now at least a year old, and read who is still active at each checkpoint. Weight the free channels first: local search, your Business Profile, referrals and buddy passes. Let paid search and social fill the gap those leave rather than replace them, and track every trial's source on the record so a cheap join that cancels in month two cannot hide inside a bigger join count. Ask any agency or vendor for joins and still-active members by source from your own software, never for leads.

How long do you have to watch a new member before you know whether a channel paid off?

HFA's summary of its paid 2026 US Health & Fitness Consumer Report says a record 81 million Americans ages 6 and older belonged to a gym, studio or other fitness facility in 2025, 26.1 percent of that age group (Health & Fitness Association, checked 2026-10-04). HFA, the Health & Fitness Association, is the trade group formerly known as IHRSA. A join is only the start: HFA's summary of its 2025 Fitness Industry Benchmarking Report, built on data from 175 companies running more than 17,000 facilities in 27 countries, says member retention averaged 66.4 percent for the year (Health & Fitness Association, checked 2026-10-04), so on average about one member in three was not kept.

That benchmark mixes 27 countries, is not a target for your gym, and says nothing about which of your channels brought the members who stayed. A cohort, the group of people who joined in the same month, needs three checkpoints to tell that story: 3 months catches the early cancellations, 6 months catches the slower ones, and 12 months lines up with the yearly retention figure above. Pull a cohort that is at least 3 months old for a first read and at least 12 months old before trusting the number.

Which free channels turn into members: local search, your Business Profile, referrals and buddy passes?

Your Business Profile is the free listing Google shows on Search and Maps. Its Performance report counts how many people clicked your call button or website link, asked for directions or started a message conversation (Google, checked 2026-10-04); none of that tells you who joined, so point the website click to a page used nowhere else and give that page's visitors a source tag in your membership software's sign-up form.

Referrals and buddy passes, a guest pass a member hands a friend for a free visit or class, are the other channel that costs little beyond the pass itself. Ask every new member how they heard about you, pick the answer from a short fixed list rather than a free-text box, and log the referring member's name when the answer is a referral. That name is what lets you later check, cohort by cohort, whether referred members stay longer than members from any other source. The worked example below walks through exactly that.

Do social posts and local partnerships bring members, or just attention?

Followers and likes are not joins. A social post can fill a class or spread word of a challenge, but unless the link in your bio or story goes to a page that tags the visit, your membership software has no way to connect a join back to that post. Add a UTM tag, a short label added to the end of a link, to every link you post, and have your sign-up form read that tag into the join record.

A partnership with a nearby physical therapy office, a corporate wellness contact or another local business works the same way as a buddy pass: give each partner its own short code or a dedicated landing page, and log which partner sent each join. Without that, a partnership that is actually bringing members looks identical in your records to one that is bringing nothing.

What does a paid click cost, and when does bidding on it make sense?

Running Google and Meta ad accounts from October 2023 to September 2025, with nearly $700K of cumulative spend between them, taught me that neither platform can tell a two-week trial from a paying member unless your front desk sends that word back. Cost per click (CPC) is what you pay each time someone clicks a search ad (Google, checked 2026-10-04). For a sense of scale, this table holds Google Ads Keyword Planner estimates for the US, October 2026:

These are estimates for the whole US, local prices differ, and none of them is what you will pay: Google describes the planner's figure as "the average amount you might pay for a click" (Google, checked 2026-10-04). Bidding on "gym near me" makes sense once your free channels are already tracked by source and the landing page the ad sends people to is used nowhere else, so a paid join never gets credited to organic search or a buddy pass by mistake. If you also sell personal training or classes, bid those terms in their own campaigns so each is judged on the members it brings at its own click price.

Between the trial and the signed membership, who follows up, and by when?

Name one person, by role, who owns every trial, whether it starts online, at the front desk or through a buddy pass. Decide a fixed follow-up window, a call or a text within a day or two of the trial's last visit, and log it as a task in your membership software rather than a note someone might forget. The trial record should carry the same source field as a direct join, so a trial that never converts still tells you something about the channel that started it.

A trial that converts fast can still cancel fast. Gym management software covers what to ask a vendor about carrying that source field from trial through to cancellation. That source field is what the worked example below depends on.

What should a marketing report show you: joins and members still paying, by source?

A report that counts leads or trial starts is reporting activity, not outcome. Why an ad account's conversion count runs ahead of what your own records show explains the usual reasons a platform's number and your membership software's number disagree, and the fix is the same here: count from your own software, not the ad account.

What to ask your agency or vendor each month:

• Joins by source, pulled from your membership software, not from its own dashboard.
• Members from each source still active at 3, 6 and 12 months, for cohorts old enough to show it.
• How it gets trial conversions from your membership software back into the ad accounts, since that is what lets the platforms' bidding learn which clicks became members.
• Cost per join by source, including ad spend and any management fee, next to cost per member still active at 3 months.

What is one new member worth, from join through month twelve?

Illustrative: the neighborhood gym below does not exist, every number in this example is invented, and none of it comes from a client. Monthly dues are $59. In the same January, 18 people joined through a buddy pass at no ad cost, and 24 joined through a social ad running a two-week free trial, at $1,200 of ad spend for that campaign, $50 per join.

A rough member value: add up the months of dues the cohort paid in its first year, multiply by monthly dues, then divide by the number of joins. Between two checkpoints, average the active count at each end and multiply by the months in between (3, then 3, then 6). This assumes members who left between checkpoints left at an even pace, which they may not, so treat it as a planning figure; your software's payment history gives the exact one.

Buddy pass: (18 + 15) ÷ 2 × 3 = 49.5, (15 + 13) ÷ 2 × 3 = 42, (13 + 10) ÷ 2 × 6 = 69, so 160.5 member-months; 160.5 × $59 = $9,469.50; ÷ 18 joins = $526.08 per join, at no ad cost.

Social ad: (24 + 9) ÷ 2 × 3 = 49.5, (9 + 4) ÷ 2 × 3 = 19.5, (4 + 2) ÷ 2 × 6 = 18, so 87 member-months; 87 × $59 = $5,133; ÷ 24 joins = $213.88 per join, minus $50 ad cost per join = $163.88 net.

The social campaign brought more people through the door, 24 against 18, but each buddy-pass join was worth more than three times the social join's net value once the months that followed are counted. A join count alone would have told the owner to spend more on the ad; the 12-month cohort tells a different story.