Having watched "The Founder", when Ray Croc was selling ice-cream machines nobody wanted he was listening to motivational records where the speaker says "the world is full of educated fools" and highlights that the one quality required is persistence.
Of course, if it's a stupid idea or a useless app then persistence is also useless and it would be a true waste of time.
But if the app is solving a problem for you, it likely will solve a problem for someone else which is why a side project will always prevail with persistence.
I always think that a side project is still a project, much like your main job. The only difference is time spent on it. But it'll be the same outcome.
For me, I find that if I work on something, eventually, my interest in it wanes and other things become more important. Unless I start making significant progress, its then only a matter of time before it fizzles out. There's only so many weeks or months I can work on something before my mind craves something else.
What definitely helps me stay focused long term is to not work on something on my own. When others are involved, then when my mind starts thinking of new things, I can set it aside and say to myself "no, stay focused, I don't want to let ABC down". Sadly when ABC is me, it doesn't work very well, but if its someone else, especially someone I care about, then it helps me stay focused. Currently on month ten of a large side project with my brother and still going strong :)
You are persistent if you keep coming back to the same task and advance it.
Let’s say that for an app, and with limited time, we can instantiate this definition as: if you keep coming back to developing your app regularly during a full year, you are persistent.
Now that we have a definition, what can you do to fulfill it?
Set time in your calendar, create habits, etc.
Rebuilding the product we ship at $dayjob, and which I enjoy working on, with a comparable feature set would probably take me ~8 years fulltime, assuming I know the whole stuff as well as I think and I don't hit major blockers (e.g. I don't do UI stuff, but most work is in the backend anyway); there is virtually no FOSS to accelerate this. A lot of the features can be dropped to get to a MVP, but I suspect I'd still be working on that for 6 month to a year, fulltime, and I will not be as easily extensible as the current incarnation. So that's 1000 to 2000h with a huge amount of technical debt and magnitudes inferior to the "competition"/$dayjob.
Could I still pull that off if I wanted to? Yes. Would I enjoy it that way? Hell no.
It's tough, shipping is only the second step.
One concern I have with a few hours a week is unless the project is really really tiny, you can begin to think that it'll never get done and it fizzles into nothingness.
For example let's say you want to make X app and it takes you 2 full time months to go from an empty folder to shipping an MVP. Let's just call it 300 hours. But if you only put in 5 hours a week (about 1 hour a day) it will take you 60 weeks to finish it. That's over a year of chipping away on something which is a very long time. You'll probably drop it well before you finish it because it'll feel like it's dragging along forever.