The Two-Thousand-Dollar Bike That Stopped Being Used

In December 2020, at the peak of pandemic enthusiasm for home fitness, Peloton Interactive briefly carried a market capitalization of approximately fifty billion dollars. The company had sold roughly two million connected exercise bikes by then, each retailing for around two thousand dollars, each generating a recurring forty-four dollar monthly subscription for the streaming fitness content that paired with the bike. The math the market was running was simple: two million bikes, each becoming a long-tailed annuity at forty-four dollars a month, growing at twenty percent a year, in a world that had just discovered the home-workout business permanently.

Three years later, in the summer of 2024, the same company's market capitalization had fallen below one billion dollars. The stock had declined approximately ninety-eight percent. The CEO who had presided over the peak had been replaced. A second CEO had been replaced. The new bike that was meant to be the next-generation hardware was delayed. The customer churn rate, formerly described as exceptionally low, had risen materially as gyms reopened and the home-workout intensity began to fade.

The math the market had been running, it turned out, had a flaw. The flaw was the customer's life.

The Annuity Assumption. The original Peloton thesis depended on a particular forecast of customer behavior. A customer who had paid two thousand dollars for the bike, the thesis went, would continue paying the forty-four dollars per month indefinitely. The friction of cancellation was high (the bike sits in the room; the implicit guilt of giving up is real). The substitute experience (a normal gym, or a video on YouTube, or jogging outside) was inferior. The fitness habit, established during pandemic isolation, was sticky. Lifetime value, modeled at five to ten years per customer, was substantially in excess of the bike's cost of goods sold. The unit economics, on this model, were extraordinary — the bike effectively becoming a customer-acquisition expense for a high-margin subscription business.

Three things, taken together, were wrong with the model.

First, the cancellation friction was lower than projected. The bike could continue sitting in the room without the subscription. Many customers, after twelve to twenty-four months, downgraded to cheaper subscriptions, cancelled, or quietly stopped logging in. The natural human pattern of new-fitness-equipment enthusiasm — six months of intense use, fading to occasional use, fading to attic — held more strongly than the venture-capital-style retention curves had assumed.

Second, the substitute experience improved. Apple, Hydrow, Tonal, and a dozen smaller competitors entered the connected fitness category during 2020-2022. The customer who had purchased a Peloton in early 2020 had, by 2023, options that did not require an additional thousand-dollar commitment. Some defected. Many simply downsized.

Third, the cost structure of the operating business turned out to be much higher than the hardware-margin model implied. The streaming content (live instructors, studio production, music licensing) required ongoing investment that scaled with subscriber count but not linearly. The customer service organization needed to support millions of pieces of expensive, heavy hardware was extraordinarily costly. The supply chain — bikes, treadmills, accessories, spare parts — generated substantial working-capital requirements that did not behave the way a pure-software subscription business would behave.

The Lesson Hidden in the Treadmill. What the Peloton experience demonstrated, in unusually clean form, was that subscription math depends critically on the assumption that the underlying experience continues to be valued at the subscribed price. When the experience is fitness — an activity for which human motivation is famously inconsistent — the assumption is fragile. When the alternative experience improves materially, the assumption breaks. The reason gym memberships have historically carried high churn rates, despite extensive industry effort to reduce them, is that gym memberships compete against the customer's own future apathy. Peloton, despite the elegance of its hardware and the strength of its content, was running the same competition. It lost, not catastrophically but predictably, and the fifty billion dollar valuation became a particular kind of monument — to the assumption that pandemic behaviors would persist, that subscription churn rates were structural rather than circumstantial, and that lifetime value could be modeled without modeling the customer's gradually returning ability to walk past the bike on their way to the door.