A new Slop University paper audits a share-housing review app's five-level landlord-responsiveness index against independently confirmed repairs, and finds a low-cost acknowledgment tool moves the score far faster than the work it is meant to track.
An independent audit of Slop University's new Landlord-Responsiveness Maturity Model finds a commercial auto-acknowledgment tool moves the score in weeks, well ahead of any confirmed repair.
A score that predicts its own ranking better than it predicts a completed repair is measuring something, just not the something on the label.
— Dr Thandiwe Solberg, Senior Lecturer and Convenor, Demo Quarter
That is the headline finding of a new paper from the School of Continuous Improvement, which built and independently audited the Landlord-Responsiveness Maturity Model (LRMM) now used by a share-housing review app to decide whether a landlord’s listing reaches a prospective tenant’s default search results at all.
Dr Solberg, Dr Bram Ntuli and Dr Sten Okwuosa tracked 1,214 listings across a metropolitan rental market for six months, setting the platform’s own five-level score against a blinded telephone audit of what tenants say actually happened to their repair requests. Landlords who took up a commercially available auto-acknowledgment tool after the platform introduced its search-visibility threshold saw their displayed score climb sharply within weeks; the same landlords’ audited repair-completion times barely moved.
“We are not in a position to say a landlord who adopts a faster reply tool intends anything by it,” said Dr Ntuli. “We can say the index currently can’t tell the difference between a message and a repair, and the search results can.”
The School frames the audit as consistent with its founding conviction that an instrument invites the same scrutiny once it starts making decisions for someone else — even one of its own design. The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.wg4448.