I recently got a haircut and one thing I’m always struck by is the inconsistency of quality. Even if I’m going to the same barber or hairdresser for successive haircuts, I still face a risk of getting a subpar haircut. That realization got me to thinking about whether something could be done to fix haircuts.
While I haven’t done the customer development to verify this, I hypothesize that hairdressers struggle to maintain quality because they don’t have reference models for each haircut. I know that my accounts of how my hair gets cut varies each time, so perhaps it’s a failure on my part to be insufficiently descriptive. That means that a single source of truth could mitigate the discrepancies in customer descriptions.
I’m envisioning a multi-part system that can turn haircuts into a data-driven exercise. First, a mobile app that uses machine vision to stitch together a 3D model of the customer’s head from photos. Second, a telescoping stand that the hairdresser can place an app-enabled phone on, at the appropriate height to capture the customer’s head. Third, the hairdresser just spins the customer around in the chair in order to render a model.
By doing this before and after a haircut, the hairdresser develops a model that they can use to ensure the quality of a haircut every time. And, of course, a customer can ask for this during subsequent haircuts if they want to update their model, in case it initially captured a bad haircut.
I also think this could give rise to smart clippers. With a reference model in place, the clippers could gather data on whether the hairdresser is on track to hit the end-state model. The app could then offer corrective advice where necessary. I’d personally love to see something like this get built, because I’m tired of subpart haircuts.
Also published on Medium.