Instead of guessing what other people may need, try creating a product that you need. Unless it’s super niche there are probably other people who need such a thing too.
The problem with that approach is that it's quite easy to make something you "need" that also isn't valuable to others. If the goal is to get paying customers, you also need signals that your solution is valuable enough to pay for.
My slight problem with that is I often have a problem say at work, but realise it is very specific to this job, this teams way of doing things. I feel that it’s worth asking around a bit before building anything.
My rule of thumb is make what you need in order to make what you needed.