Bob Lutz was CEO of Chrysler at the time, and they were market testing different designs for the next Dodge Ram pickup truck (1990's). They went with a design that the majority of people didn't prefer, however those who did prefer it, loved it.
They figured by going after the most loyal and vocal buyers, they'd have the most success.
I always feel like you should just pursue the most passionate 10% of users who are excited about your product, as that enthusiasm is contagious and will eventually persuade everyone else.
I remember when those came out. My mother, who owned a previous generation Dodge Ram hated the look. A decade or so later, she bought a Dodge Dakota with similar styling cues. She didn't hate it enough to turn down the right combination of price and features.
This is probably a good approach to most markets that are relatively commoditized (early '90s GM, Ford and Dodge full-size pickup trucks were roughly interchangeable): find some way to make yours stand out that a fraction of users are really passionate about, then double down on it.
The 2nd gen Ram design aged terribly compare to Ford or GM's mid/late 90s departure from brick-like trucks. You just don't notice how terribly they aged because they fell apart and rusted out faster
That depends a little on your market position. Subaru did the same thing, but that strategy probably won't work as well if you're in the Corolla/Civic part of the market.
They figured by going after the most loyal and vocal buyers, they'd have the most success.