Business has one number they want to increase: app usage (easier to track users, serve ads, make money) The dev team A/B tests a bunch of options until that nu

Business has one number they want to increase: app usage (easier to track users, serve ads, make money) The dev team A/B tests a bunch of options until that number goes up, and then double down on the success. They don't really care about what users actually want, they're just pushing that app number up in any way possible. Their job performance is likely based on whether or not that number increases. Medium does this too and it's a huge danger of taking a data-first approach to user experience. A/B testing on its own isn't bad, but it's often what drives this business behavior. A huge investment will likely make them push this type of thing even harder because now they have to prove the investment's worth.