A new Slop University research poster audits a household meal-kit subscription's per-meal premium against the grocery shop it replaces across three tiers, finding a randomised cost-disclosure mailer produces no detectable change in cancellation.
A new research poster audits 214 subscriber households across three box tiers and finds an itemised cost breakdown changes no one's mind about cancelling
The box doesn’t get cheaper the longer you keep it. It just gets harder to remember you’re still deciding to.
— Dr Marek Solheim, Lecturer, School of Emergent Priorities
Slop University has published a new research poster comparing what a household meal-kit subscription actually costs per meal against the grocery shop it replaces, and testing whether telling subscribers the gap changes anything. The poster, from the School of Emergent Priorities, audited 214 subscriber households across three subscription tiers over nine months, pricing a matched grocery-equivalent basket alongside each box.
The subscription carried a per-meal premium of 36 to 60 percent over its grocery-equivalent basket at every tier, with the family box — marketed as the discount option — carrying the smallest absolute gap and the same relative one. Halfway through the audit, researchers began mailing half the panel an itemised monthly cost breakdown; by the ninth month, cancellation among mailed and unmailed households sat within a percentage point of each other.
“The premium isn’t the interesting part,” said Dr Renke Sabel, Senior Lecturer and Convenor of the Indicator Commons. “The interesting part is that showing someone the number changes nothing about what they do with it.”
Exit interviews suggested why: most subscribers had already estimated a premium in roughly the right range before anyone told them, mailer or not. What moved was their confidence in the guess, not the decision resting on it. The finding lands in a portfolio the School has spent the year building — household instruments that keep functioning long after anyone reads them closely.
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.fy3xp4.