I find hundreds or thousands of complaints. Many are legitimate bugs, and some are user stupidity ("I need to check my mums email but she won't give me her password, can you let me in anyway").
For those which are legitimate bugs, I know, through user metrics, that they only affect a tiny fraction <0.1% of the users. I, and the rest of the engineering team have to prioritize our team between fixing these bugs for the 0.1%, and making the product better for the 99.9%.
It can be a tough choice, but when you've decided that it's time to move on and develop new features, hearing about individual instances of rare bugs is no longer useful. We just aggregate how many users are impacted by each significant bug, and from time to time quash the top ones.
Manually editing a database entry for a single user is no longer privacy-justifiable, so basically the only fix we can do is to fix the bug for all users at once. And if the bug is only affecting 15 out of 300,000,000 users on a product that makes 1 cent per user per year, I can't afford to spend more than 30 seconds on it really, yet most bugfixes are at least a days work.