In the past I worked for a large company with data analytics including user tracking. Since I've seen how much (unneeded) data was tracked for any user.
As I was checking compatibility of sites and apps with all kinds of browsers and devices I could check which data can be blocked but still have the site or application still work.
I couldn't see any explicit user information, but that wasn't really needed. If you combine multiple sources you could determine a user based on their meta information (like device+location).
Since I'm more aware of sharing information with others and block a lot of data and not use many apps (apps are riddled with tracking and harder to block without some proxy).
Next to this I learned that HTTPS isn't as safe as presented. Yes it encrypts your communication, but the network provides the certificate and can access the data if needed.
The only safe way of using HTTPS is by using certificate-pinning (
https://wiki.mozilla.org/SecurityEngineering/Public_Key_Pinning), else a 'safe' proxy can be used which is under control by the network you're using. Certificate pinning forces the server to be using your certificate, not just any certified server. (Obviously HTTPS can eventually be decrypted, so safe is a relative term here)
I also once used HTTP 301 or 308 redirect on a 'safe' proxy to force iPhones to use my own test server instead of a commercial website. If you switched to a regular wifi/mobile network the redirect would stay in place without the user knowing.
As I'm just an engineer I assume many of these tricks can be used to take attack unsuspecting users very easily.
I also assume that if a big organization (like a government) wants to hack any of you devices they can, all devices can be hacked but some are more easily hackable compared to others.