A new Slop University paper randomises a $20 booking deposit across the 38 salons of one metropolitan hairdressing chain and finds no measurable change in the no-show rate, alongside a small fall in bookings and rebookings.
A cluster-randomised trial across 38 salons puts the difference at a third of a percentage point, against a before-and-after figure ten times larger
Two numbers came out of the same 38 salons over the same twenty-one weeks. One of them reached a decision four weeks before the other existed.
A paper published today by the School of Continuous Improvement reports a cluster-randomised trial of a $20 booking deposit at a metropolitan hairdressing chain, with nineteen salons taking the deposit and nineteen continuing to book on trust. Across 61,430 booked appointments the no-show rate was 7.6 per cent under the deposit and 7.9 per cent without it, a difference inside the margin the chain’s own operations team had described as worth having. The same records, compared before and after within the deposit salons alone, return a fall of 3.2 percentage points, a reduction the paper finds present in full at the salons that never took a deposit.
An evaluation carries two dates: the day it answers the question and the day the question was last open. The interesting work is in the gap between them.
— Dr Dagny Okafor, Lecturer and Convenor, Evaluation of Evaluation
The trial applies the School’s evaluative capability to a setting where a concurrent comparison is rarely available, and pairs the randomised estimate with the observational one an organisation’s reporting cycle produces on its own. Dr Ingeborg Nwachukwu, Lecturer and Convenor of Capability Sprints, noted that “everyone involved acted reasonably at every step, which is a large part of why the case is worth teaching from”. The University regards work of this kind as central to the anticipatory posture the School has been building across its evaluation programs.
The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.1jjfkf.