So they did an analysis and determined that only 6,000 or so of the products they carried sold in any significant amount. The other 12,000 were niche items that sold in small quantities and made little money.
Before they nixed the other 9,000 products though, the company decided to make one additional analysis, of who bought what in what combination. And it discovered all those little items were being bought in small quantities by people who bought large quantities of the big items.
The analysis further determined that if the supermarket nixed those 9,000 niche products, it would no longer be the "one stop shop" for its customers. Those customers would only be able to buy some of what they needed at the supermarket, and would need to buy the rest elsewhere. If a competing supermarket carried all of what a customer needed, that customer would eventually learn to do all his shopping at that supermarket to save the hassle of needing to journey to 2+ places to get what he needed.
After this final analysis, the supermarket chain decided to keep almost all the products it kept on its shelves, eliminating only 200 or 300 products, instead of 9,000.
It was not so much that the supermarket was very nice and charitable to niche spices and exotic Asian sauces and imported curries and whatnot.
Rather, it was that the supermarket realized that if its customers could not find everything they wanted on its shelves, they would eventually find everything they wanted on a competitor's shelves instead.
In YouTube's case, no good competitor presently exists. But a big part of the reason why is because YouTube has sucked all the oxygen out of the video hosting and streaming niche by being a one-stop shop. If it stops being that, there will be a lot more oxygen in that niche again.
(that said, I do not think demonetizing videos with low view counts / subscriber numbers constitutes injecting oxygen back into the market. "Not letting" niche video creators user their platform, however, probably would)
