Manifesto
Why one outfit beats twenty options
In 2000, Sheena Iyengar and Mark Lepper ran an experiment at a Menlo Park grocery store. One day, they offered shoppers a choice of six jams. Another day, twenty-four. The wider choice attracted more attention; the narrower choice produced ten times the purchases.
The lesson has been repeated to the point of cliché: more options paralyse. But it has also been ignored, almost universally, by every fashion app on the market.
The default is a grid
Pinterest is a grid. Whering’s outfit board is a grid. Instagram Shopping is a grid. The grid is a default because it transfers the cost of choosing onto the user, which is cheap for the platform and expensive for the human standing in front of an open closet at 7am.
The grid is also where shopping lives. Every fourth tile is something you could buy. The grid is, in the end, optimised for the platform’s revenue, not your morning.
What “one outfit” requires
For a recommendation to be confident enough to stand alone, the system has to know what you have, what’s clean, what fits the day, and what you wore yesterday. If any of those reads is missing, the recommendation is a guess, and a guess is a grid waiting to appear.
This is why BURS reads the day before answering. One outfit, from your closet, for this Tuesday — that is what the model is trying to deliver. Not twenty. One.
When one is wrong
The cost of being wrong with one outfit is, in theory, higher than the cost of being wrong with twenty. In practice it is lower. With twenty options, “wrong” is invisible. With one, “wrong” is correctable in a sentence: make this warmer, softer palette, something for 7pm dinner. The chat is the safety net, and the safety net is what makes the one-outfit design viable.