If people opt out of what you're doing to them as soon as they're made aware you're doing it, then perhaps that's a sign that you shouldn't have been doing it in the first place. I don't even believe that 25% of users actually wanted tracking. I think a lot of people just blindly tap buttons to make annoyances go away, and these guys just hit the wrong button.
As soon as ad supported free services start shutting down because of ad blocking and lowered clickthrough rates on ads because of targeting being blocked, most people will probably start changing their minds. (The alternative being maintaining 10-20+ paid subscriptions.) For now, all this change means is that users who are opted-in to tracking are subsidising those who aren't.
1) People want the services more than they value their privacy. Maybe they'll just not use the service if they can't use it without tracking
2) That invasive tracking is required to sell ads. The media industry made billions (trillions?) of revenue from ads before tracking became a thing.
3) That platform ads are the only way to make services that are free for consumers. For example, Vimeo offers an ad free video delivery service that the content creator pays for. If Youtube was no longer free, maybe content creators would just pay for content delivery instead of having consumers indirectly pay for deliver with ads. Content creators have no issue selling ads / sponsorships without any tracking whatsoever. The result would be the same as now (content free for consumers) only that now non-targeted ads would pay for everything.
4) And finally, you are assuming that targeting via tracking actually works well enough to make it worthwhile. From what I've read, ad targeting is nowhere near as good as Facebook et al would have advertisers believe. Maybe invading your users privcy just doesn't make such a big difference in the end.