Idea validation: a database of mechanical keyboards

Would this work? I think I know the ingredient necessary - where to get data, how to present it, and I'd certainly be able to build the thing. But I'm thinking of this as one of my first times dipping my toes into a side project that "makes money" and working back from the 2-8% commission I would get if people click through and buy keyboards, it'd require people to buy ~100 keyboards/month in order for it to net $1k MRR.

Are affiliate-based projects just something to be avoided? Am I setting the bar too high?