One member in three leaves every year and the market is growing anyway — that's the trap

The market is genuinely growing: 81 million Americans held a membership in 2025, 26.1% of the population aged six and over, up 5.2% year on year. Net membership growth ran at 5.5% and median revenue growth at 9.9%. Underneath those numbers, annual member retention is 66.4%, which means roughly one member in three leaves every year and the acquisition machine is running at full speed to produce a net figure that looks like progress. The published cost of that machine is $67.36 a lead in health and fitness and $44.26 in sports and recreation.

A facility can grow membership 5% in a year while replacing a third of the people in the building, and the growth number is the one that gets reported.

1. 01

   Net growth is the number that hides the problem it is caused by

   5.5% net membership growth against 66.4% retention means gross acquisition is doing enormous work to produce a small net figure. A facility reporting only the net number has no visibility of the cost of standing still, and no reason to fund the cheaper fix.

2. 02

   The cohort everyone markets to is the cohort that leaves

   Gen Z aged 18–24 has the highest penetration of any group at 35.5% and the highest churn at 54.42%. The 65+ cohort is the fastest growing at +8.6% and churns at 26.48%, roughly half the rate. Acquisition budget almost always follows the first number and ignores the second, which means it is buying the most expensive members to keep.

3. 03

   The dormant paying member is close to extinct

   The share of members not visiting at all fell to an all-time low of 4.6%. The old cushion — people who paid and never came — has largely gone. People who stop attending now cancel, which removes the lag that used to hide a retention problem for a couple of quarters.

4. 04

   The two dollar figures everyone quotes do not exist

   Neither member acquisition cost nor member lifetime value is published credibly for this industry. The only defensible comparison available is the published cost per lead against the facility's own measured member value, and until the second is measured the first is a number without a denominator.

5. 05

   A single retention percentage across the whole membership describes nobody

   With churn running from 26% to 54% across age cohorts, an aggregate retention figure is the average of two different businesses. Retention has to be read by cohort and by acquisition source or it cannot be acted on.

**81 million** US gym members in 2025 · annual retention **66.4%** (≈33% churn). *(Cite 66.4%, not the legacy 66.4%.)*

The model funds the first ninety days of membership before it funds the next lead. Instrumentation comes first: join date, first-visit date, visit frequency by week, cancellation date and stated reason, all joined to acquisition source, so retention can be read by channel and by cohort rather than in aggregate. Onboarding is then defined rather than assumed — a fixed number of touches across the first six weeks with at least one human contact, and a first visit booked at the point of joining rather than left to the member. Acquisition scales last, weighted toward the cohorts with the published retention advantage, and is measured on members retained at ninety days rather than on joins.

1. 1

   Weeks 1–4

   Retention instrumentation

   Join, first visit, weekly frequency, cancellation and reason joined to acquisition source

   Owner: OmniFlow + facility

2. 2

   Weeks 4–6

   Cohort baseline

   Retention curve by age band and by acquisition source, not in aggregate

   Owner: OmniFlow

3. 3

   Weeks 6–12

   Onboarding programme

   Fixed touches across six weeks, one human contact, first visit booked at join

   Owner: Facility + OmniFlow

4. 4

   Week 8 onward

   Cancellation capture

   Stated reason recorded on every cancellation, coded to a fixed list

   Owner: Facility

5. 5

   Month 4 onward

   Acquisition rebuild

   Campaigns weighted by cohort retention, measured on members retained at 90 days

   Owner: OmniFlow

6. 6

   Month 12

   Review

   Retention by cohort and source, cost per member retained at 90 days

   Owner: OmniFlow + facility

The model reports cost per member retained at ninety days alongside cost per lead, and retention by acquisition source rather than as a single facility number. Because no credible published figure exists for member acquisition cost or member lifetime value, the published cost per lead is used as a ceiling on media spend rather than as a target to optimise toward, and the facility's own first two quarters become the only valid basis for comparison. Retention is always reported by cohort, because a single percentage spanning an eighteen-year-old and a seventy-year-old is an average of two businesses that behave nothing alike.

Modelled. Inputs: Health & Fitness Association, 2025 Fitness Industry Benchmarking Report (175 companies, 17,000+ facilities); Health & Fitness Association, 2026 US Health & Fitness Consumer Report (~18,000 US residents); WordStream by LocaliQ, Google Ads Benchmarks 2026 (13,474 US search campaigns, April 2025 – March 2026). Modelled outputs are not a forecast or a guarantee of results. Cited: Health & Fitness Association *2026 US Health & Fitness Consumer Report*. Figures refreshed 2026 against HFA 2026 (81M US members, 2025).

Every figure below is published research, not a client result. They are the inputs to the arithmetic above, listed so it can be checked rather than taken on trust. The bracketed number points to the full citation at the end of this page.

These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.

Each figure on this page, the system it is read from, and the definition and window it is measured over.

Not applicable — this is a modelled engagement. Its weakest input is that the retention and churn figures are drawn from operator-reported data across a base weighted toward large multi-site chains, whose contract structures, cancellation friction and member mix look nothing like an independent single site. An independent facility's retention curve may be materially better or worse and the aggregate cannot tell it which. The instrumentation phase exists to establish the facility's own curve before a single dollar of budget moves.

A modelled capability, not a client account. Figures illustrate what the model produces and are labelled as modelled wherever they appear. See all our work

Full citations for everything cited on this page, with the sample and period each study covers, so you can go and read the original.