Often the ideas come from some obstacle or problem that you've experienced.
E.g. during the COVID almost nothing worked and I was stuck at home curious if there is still anything happening around me? And I literally wanted to see on the map if there was anything in my neighborhood at all. But none of the existing solutions worked for me. So I decided to make my own. So I've created https://wiwo.world/map. There you can find things that are happening nearby.
With me the problem is more with the consistency of the executions rather than the idea itself.
As I wrote in another comment I often abandon a project for the next one.
Honestly, I do thing the execution is much more difficult and important than the idea. For example in our above project we have tons of ideas what other features to include but we want to make it with the active feedback by the users and we have a huge struggle to find any early adopters, mainly because we have never searched for such before.
One other controversial but possible approach is just to ask AI for ideas.
Or maybe you can take a look at all the existing products and see how you can improve one of them in your own way.
Unfortunately however, as I said above, most of the time, people fail when it comes time for the execution.