What 86 Behavioural Science Studies Say About Where to Focus Gym Retention Efforts

In Peloton’s 2019 S-1 filing, the average Connected Fitness Subscriber did 11.5 workouts a month, and the twelve-month retention rate sat at 95%. Both figures had been climbing year over year through the prior reporting period. The same filing put the cause on the page in plain language: engagement is the leading indicator of retention.

Engagement and retention are the same line on the dashboard, separated by time. A twelve-month retention figure of 95% gets set in the first three weeks of the relationship, before any retention strategy has a chance to operate. By the time a quarterly review runs, the number is already locked in by an engagement pattern that has been in place for months.

Most gyms run the inverse pattern. Their onboarding is a five-minute tour, a welcome email, and a name on a membership database. Twelve-month retention then runs as an automation sequence chasing members who were never properly engaged in week one. Spend gets concentrated at the end of the lifecycle, where the discount curve has already settled, and starved at the beginning, where it could still bend. The cognitive economics of why this fails sit in a 2025 meta-analysis published in Management Science.

What 86 studies on self-control found

In 2025, three economists published a meta-analysis in Management Science covering 86 separate studies of quasi-hyperbolic discounting. The model captures present bias as a single number between zero and one, where one is patience and zero is full impulsivity.

Across those 86 studies, the present-bias parameter for money averaged 0.938. Once the same studies switched to non-monetary rewards like effort and real consumption, the parameter dropped to 0.750. Selective-reporting tests turned up nothing, which is unusual for a literature this size.

Translated for a gym owner: when people are asked to wait for future money, they keep around 94 cents on every promised dollar. The same people asked to hold out on future effort keep about 75. That margin is the practical difference between a retention investment that compounds and one that drains.

Why this happens in your gym

A new member walks in carrying a temporary state most retention strategies never bother to use. Their decision is fresh and motivation is high, which leaves the discount on near-future effort sitting at the mildest point of the entire member lifecycle. Week-one prospects will agree to commitments they would politely decline in week ten, made by the same person under different cognitive conditions. That window is short and measurable, and it closes faster than gym onboarding sequences are typically built to handle.

Onboarding is the one period where the brain registers effort commitments at favourable terms. A first session gets booked while the member is still at the desk. The second is usually pencilled into the calendar before the member leaves the building. Within ten days, the schedule starts to operate as a default rather than something the member negotiates with each week. Habit is the long word for what’s happening, but the cognitive mechanism behind it is more elementary. With the discount curve temporarily flat, the member is approving future-effort commitments at a price the future self will inherit and accept, because by the time the future self arrives, the deal is already a default.

After thirty to sixty days the curve returns to baseline, and the member’s working relationship with the gym becomes whatever was established in that opening window. The behavioural economist David Laibson worked out in the 1990s that humans don’t run a single discount rate for the future. They run two, and the near-future rate is far more punishing than the far-future one. Cheung, Tymula and Wang’s 2025 meta-analysis added a finding the original model didn’t predict: the punishing near-future rate gets sharper still when the future cost is effort instead of money. By month three, both rates are fully operational, and any retention intervention has to work against them rather than with them.

Win-back campaigns are designed to operate in the worst possible window of the entire lifecycle. By the time a member has lapsed, the discount curve has fully steepened, and the relationship with the gym has gone dormant. The brain at that point is no longer accepting new commitments at any price the campaign can offer. A great win-back campaign is fighting a much more advanced opponent than a five-minute onboarding investment had to. The retention budget that lands there fails on timing alone, regardless of how well the campaign itself is built.

Where this goes wrong

The flat-curve window does not give the operator permission to push a vulnerable new member through aggressive scheduling and high-pressure habit-building. That’s the same exploitative move under a different label, and the cognitive science doesn’t make it less so. Members in week one can still be overwhelmed, and over-loaded onboarding produces a different cancellation pattern than under-loading does, sometimes a faster one. The realistic use of the window is to set up sustainable habits at a pace the member can carry forward into the steeper part of their own discount curve.

Some onboarding spend is wasted regardless of how it’s deployed. Members who joined for a one-off goal (event prep, holiday lead-up, a January resolution) were never planning to form a habit. They were buying access to a defined-window service, and the cognitive economics of long-term retention don’t apply to them. Telling the difference between a habit member and a service member is the operator’s call, and the meta-analysis doesn’t make it. For the habit cohort, though, the leverage point is the first 30–60 days and almost nowhere else.

Measurement is the third thing that goes wrong. An onboarding investment never lands on the same dashboard as a win-back campaign, and the comparison the operator runs is therefore between two metrics measured on different timescales. Win-back has visible short-term outputs like open rates, reactivation visits, and returned members. Onboarding has the retention number months later, by which time the cause is invisible and the credit gets assigned to whatever automation ran most recently. Most gyms cannot see the leverage of onboarding because their reporting cadence isn’t built to detect it.

The question

What does the first 30 days of your member’s experience look like, and how much of your retention budget is spent there?

People also ask

When should gyms invest in member retention?

The first 30 to 60 days of membership is the window where retention investment compounds. A new member’s willingness to commit to future effort is temporarily flat at sign-up, propped up by recent decision-making and high motivation, and that same willingness decays steeply across the following two months. Behaviours, schedules, and routines locked in during the opening window become the member’s default for the rest of the relationship. Investments made later in the lifecycle face a steeper present-bias curve, and the returns shrink fast. Peloton’s 2019 S-1 disclosed an 11.5 monthly workouts average and a 95% twelve-month retention rate, with management identifying engagement as the leading indicator of retention. Both of those numbers are downstream of an early-cycle engagement pattern that was set in week three, well before any late-cycle retention activity could matter.

Why don’t win-back campaigns work for gyms?

Win-back campaigns target members who have already moved into a steep present-bias state, where any new commitment to gym effort is heavily discounted before the campaign even arrives. The 2025 Management Science meta-analysis on quasi-hyperbolic discounting found that present bias for non-monetary effort averages 0.750 across 86 studies, compared with 0.938 for money. A lapsed member offered a return-to-gym campaign is being asked to commit to future effort at the worst possible point on their own discount curve, and conversion rates reflect that. Win-back can produce some short-term reactivation, but the underlying retention economics are unfavourable, and the same budget tends to produce more compound return when it’s redirected into early-cycle onboarding instead.

What is gym retention onboarding?

Gym retention onboarding is the structured set of actions taken in a member’s first 30 to 60 days, designed to establish sustainable engagement before the present-bias curve returns to baseline. Components vary across gym formats, but the goal is the same: use the temporarily flat effort-discount window to lock in scheduling patterns, social connections, and skill milestones that the member’s future self will inherit as defaults rather than negotiate as new commitments. Good onboarding doesn’t push members harder than they can carry, and it isn’t a tool for cramming aggressive commitment extraction into week one in place of gradual habit formation. The window is short and the leverage on it is asymmetric, which is why the operator’s most consequential retention work happens here and almost nowhere else.

---

By the time a member is lapsing, retention has already happened. Most of the work happened in week one.

The timetable is one place retention spend quietly leaks. When a class goes stale, you can relaunch a tired class rather than let it fade off the schedule.

If you want a behavioural-science walkthrough of where your retention budget is currently producing diminishing returns and which parts of it could compound if redirected, book a free 30-minute chat.

References: Cheung, Tymula & Wang (2025), A Meta-Analysis of Quasi-Hyperbolic Discounting, Management Science. Peloton Interactive Inc., Form S-1 Registration Statement, U.S. Securities and Exchange Commission, August 2019. Related: The Forever Fallacy: Why Gym Members Really Cancel.