I'm based in The Netherlands.

A previous job made us track our hours so that we could "improve estimates," but mostly it only made people unhappy.

We ask all of our employees and contractors to track time against tasks. Yes it can be a pain but it helps prevent people from spending way more time on a task than there was budget for.

This is the current biggest mistake in the industry, imo.

The schedule may still slip, since estimating time is very difficult, but at least some of the time it will be correct (or estimates will get padded until things happen on time). Either way, it's more useful than giving up on the problem.

Instead teams should focus on delivering value incrementally, then using metrics over those increments as a measure for future increments.

Time estimating is broken and inaccurate. But that doesn't mean it's unnecessary, or that one can just give up on it. Teams should focus on delivering value incrementally, then using time taken as a metric (among others) over those increments as a measure for future increments.

"How many points did you spend on that story last week?" is not something anyone is going to accurately answer (and it doesn't mean anything, anyway).

But being able to look back in May and say we logged 32% of all development hours on bugs is a very useful metric for me the business owner / dev lead.

(In addition we don't do story points on bugs. I am so far away from the by the book agile thing, I don't even know if this is correct or not anymore).

Time based software development estimates before the magical "story points" system was in vogue.