[nonameiguess]

I don't think you're taking this the way it was intended. The person who posted is a moderator who does not have a profit motive. Reddit doesn't distribute profit to subreddit moderators (as far as I know?). But even without a profit motive, he is still left without any simple way to evaluate the quality of submissions and decide what should be promoted and what should be killed. The point being "engagement" is often used not because of profit motive, but even in the absence of profit motive, it is a good default metric that is easy to measure in trying to figure out what your community actually wants to see.

He's not asking what motives you should have if you don't have a profit motive. He's asking, even in the absence of a profit motive, what better measure of content quality is there than how much the community engages with it? You gave an answer. In communities you ran, you just asked yourself if you were happy with them. That's an acceptable answer. But assuming someone who runs a community wants the community to be happy as well, that is a lot harder to assess, as you no longer have direct access to their mental states the way you have to your own. Now you're left again with the need to estimate their happiness via some proxy measure. If not engagement, what should that be?

I'm not saying there is no answer and all communities should throw up their hands and either just use engagement or use no measure at all, but I do think it is legitimately a hard question.