ClassPass Variety-Seeking #

Key Result: 94% of ClassPass bookings were made at venues new to the user, while studio partner retention held at 96% - the signature of a business built on variety-seeking rather than against it (company-reported). BS-0059

Background #

The traditional gym membership rests on a behavioral assumption: buyers will commit to one venue and one routine, and repetition will harden into loyalty. For a large share of the market, the assumption is simply false. Many exercisers are variety-seeking. Boredom drives their dropout, and commitment demanded before experience produces regret rather than attachment. The industry’s standard response has been to fight the preference with long-term contracts, cancellation friction, and loyalty incentives, which converts a behavior mismatch into churn.

ClassPass made the opposite bet. Instead of asking variety-seekers to become single-gym loyalists, it sold the variety itself: one subscription granting access to classes across many studios and formats. From 2012 to 2018 the model evolved, most visibly in the 2018 shift from an “unlimited” plan to a credit system, but the underlying behavioral thesis never changed. The company was acquired by Mindbody in 2021.

What actually drove retention #

The behavior ClassPass needed from subscribers was “book and attend varied fitness classes,” and the booking data confirmed that variety was the product, not a transition phase: 94% of bookings were made at venues new to the user. Subscribers were not sampling their way toward a favorite studio and settling down. Exploration was the steady state, week after week.

Three design choices turned that stable preference into a durable business:

• Aggregation made variety operationally cheap. A single subscription across many studios means trying something new costs no extra contract, negotiation, or membership decision.
• Discovery and low booking friction kept the behavior flowing. The model is strongest exactly where classes are easy to find and booking takes seconds; every added step between curiosity and a confirmed spot suppresses the core behavior.
• Credits priced variety sustainably. The unlimited plan matched the behavioral thesis but strained the economics of repeat usage and studio capacity. The 2018 credit model preserved the variety-seeking behavior while restoring economic discipline, and the company reported its “biggest months of growth” after the change (founder interview).

The supply side confirms the fit from the other direction: 96% of studio partners stayed on the platform, indicating that the marketplace worked for the businesses fulfilling the behavior as well as the consumers performing it.

Behavior Fit Assessment #

These ratings are analyst examples of a Behavior Fit Assessment, not direct measurements. The front-matter Medium scores summarize the mixed ClassPass subscriber population; the High assessment applies only to variety-seeking subscribers in markets with sufficient supply and access. For “book and attend varied fitness classes,” Dispositional Fit splits sharply by segment: high for people with a relatively enduring preference for novelty and exploration, low for people who prefer one familiar gym and a stable program. Labels such as “explorer” and “routine optimizer” are shorthand for those preference patterns, not the basis of the score. Capability Fit is medium because physical conditioning, mobility, and class-specific movement skills vary across subscribers. Context Fit is medium because studio proximity, class supply, transportation, schedule availability, and booking access determine whether the behavior has a viable external setting. ClassPass did not try to raise fit through persuasion; it selected the segment and the markets where fit was already high and built the model around them - a clean example of behavioral selection driving Behavior Market Fit.

Results #

• 94% of bookings were at venues new to the user, confirming variety-seeking as the dominant behavior pattern (company-reported). BS-0059

• 96% studio partner retention rate, indicating the model worked for the supply side as well (company-reported).
• 62% trial activation rate when incentivized, high for the fitness category (company-reported).
• The shift from unlimited to a credit-based model produced the company’s “biggest months of growth” by aligning economic constraints with behavior reality (founder interview).
• ClassPass was acquired by Mindbody in 2021, validating the marketplace model at scale (press-reported).

Limitations #

Partner and retention statistics vary by cohort and studio mix; treat them as directional signals, not universal benchmarks. Variety-seeking is heterogeneous across segments, and the model works best in dense urban markets with high studio supply - the same numbers should not be expected where partners are sparse or schedules rigid. Company-reported metrics may not reflect all cohorts, and partner retention likely differs by studio type and geography. Finally, the move from unlimited to credits was partly forced by unsustainable unit economics, not purely behavioral insight, so the behavior thesis and the business-model correction are entangled in the growth story.

Lessons #

1. A stable preference is not a stage. Variety-seeking does not mature into single-venue loyalty, and business models that wait for that maturation buy churn. Treating the preference as behavior reality opened a market the commitment-based industry had structurally ignored.
2. Design the model around the behavior, not the ideology. The gym industry’s commitment ideology turns variety-seekers into cancellations; ClassPass monetized the identical trait by making exploration the product. Same population, opposite economics.
3. Behavior fit and unit economics must both clear. The unlimited plan fit the behavior but broke the business; credits fit both. A model that enables the target behavior while losing money on every performance of it is not a strategy.

Sources #

• The Business of Fitness: How ClassPass Co-Founder Payal Kadakia Built a $1 Billion Company (Inc., 2017)
• ClassPass: navigating the challenges of building a sustainable marketplace model (Harvard Digital Initiative)
• ClassPass introduces credits (TechCrunch, 2018)
• Evidence Ledger: BS-0059