[joemi]

I'm not entirely sure what you've described is the case. I wouldn't be surprised if in some cases the for-the-worse changes aren't done when a site is at the top of their game, but are instead reactionary attempts because they noticed downward trends for sign-ups or engagement or whatever that aren't necessarily apparent to the users yet. People don't necessarily need a good reason to stop using something besides they don't think it's cool anymore.

I'd love to hear from the decision makers behind such changes, whether it was an attempt to improve an already-good situation or if it was an attempt to correct a declining situation.