A new Slop University paper scores forty boutique fitness studios against a class-waitlist maturity model and finds the resulting quadrant explains almost nothing about how full classes actually run.
Scoring forty boutique fitness studios on a waitlist maturity rubric found the score no better predicted class-fill or no-show rates than chance, and dropping one axis changed nothing.
A maturity model built to tell a boutique fitness studio how sophisticated its class-waitlist text system is turns out to say almost nothing about whether its classes actually fill. A new Slop University paper from the School of Continuous Improvement scored forty independently run studios against the two-axis rubric vendors already use — automated spot release against staff-triggered release, visible queue position against a black box — then checked each studio’s quadrant against fourteen weeks of its own front-desk scanner data.
A studio can score full marks on both axes and still run classes at the same fill rate as the studio next door that scores zero on both. The rubric is measuring something about how the system was built. It isn’t measuring what happens to the seat.
— Dr Torun Ezeigwe, Senior Lecturer and Director, Master of Applied Measurement
The gap held everywhere the team looked. Mean class-fill rate sat within three points across all four quadrants, including the “mature” one vendors hold up as the target state, and self-reported capacity figures ran ten to fourteen points ahead of the audited number regardless of quadrant. Reclassifying every studio on automation alone, with transparency dropped entirely, reproduced the same flat result.
Dr Thandiwe Solberg, the paper’s co-author, said the finding fits the School’s long-running interest in what a classification system is actually for: “we keep finding instruments that are internally tidy and externally silent. This one joins a fairly long shelf.”
The School treats the result as a caution against building its next instrument on legibility alone; a rubric that earns consensus among its own raters isn’t the same rubric that predicts an outcome nobody consulted it about.
The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.szpgft.