A new paper from the School of Continuous Improvement interviews 32 rideshare riders about how they relate to the app's upfront fare estimate once it has run high, finding their coping habits hold steady even after being told the quoted price is informational only.
A 32-rider interview study finds screenshotting, mental rounding, and simply not looking survive being told the quoted price was never a promise
Ask a rideshare rider what they do with the price the app shows before they book, and most describe a habit, not a decision: a screenshot taken from routine, a number quietly rounded up, a glance that stopped happening months ago. A new paper from the School of Continuous Improvement names each habit and traces where it comes from.
Two researchers interviewed 32 riders, working back through each participant’s fare history for any trip where the final charge exceeded the quoted number. Riders sorted into four habits: screenshotting every quote, privately rounding the number up, checking only sometimes, or no longer checking at all.
The habit tracked how often a rider had been caught out before. Screenshotting was the default among riders recalling two or more fare surprises, and the exception among riders recalling none. Partway through each interview, researchers disclosed that the app’s own terms label the estimate “informational only, not a guaranteed price,” then asked what participants would do differently next time. Almost nobody changed their answer.
A rider who screenshots the quote isn’t waiting for the number to be wrong. They’ve just stopped assuming it will be right.
— Dr Mirela Hanke, School of Continuous Improvement
Co-author Iben Chikere, a Lecturer in the School of Emergent Priorities, said the finding fits a pattern the two schools keep meeting from different directions: “an institution can relabel a number as many times as it likes; the person relying on it has usually already worked out what to do about it.”
The paper extends the School’s wider interest in what happens once a measurement stops being trusted at face value.
The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.jugfj6.