How do you find or validate an app/SaaS idea, to the point where you actually decide to go build it?

I hadn't expected so many people to comment on my previous article, and I also had in-depth discussions with a number of developers.

A very interesting question：
How do you discover or validate an app or SaaS idea before code?

or Where does the idea come from that makes you commit to building a new app or SaaS?

Or: what actually pushes you over the line into starting a project?

And here's the more interesting version of it — SaaS and apps are already fiercely competitive, and it's only getting worse, so how do you find a good idea at all?

A wrong idea is expensive — it burns money, time, energy, and people.

I've shipped a lot of SaaS and apps by now, but there are still stretches where I honestly don't know what to build next. I just sit in front of the screen, staring, scouring the world for an idea.

Some possible approaches:

1. Future profit data. The most direct proof an idea is right — and impossible to get. If you have it, please @ me relentlessly, haha.

2. Having talked to a real slice of the target users for that idea. This is really doing the distribution before the building. The catch: research can't guarantee the idea will succeed either.

3. The developer observes and thinks about the world, and has a flash of insight. These people are frightening. That level of perception isn't ordinary — they'd succeed at anything.

I saw someone make the point that we're all too similar to each other, so our ideas end up heavily overlapping. His point was: if you want a good idea, you have to become a different kind of person in real life — go push yourself to extremes, for instance. I forget who said it, but there's something to it.

4. The developer hits a problem themselves, and goes and solves it.

5. Just ask AI.

And so on.

My own experience: some of my ideas came from daily life, some straight from asking AI — neither worked well. So I started combing through app reviews, reading Reddit, checking search volume. That process drove me crazy, so I built RightIdea to do it for me.

It can't guarantee the ideas it surfaces and validates are right, but that doesn't make it worthless either — call it limited value.

It offers a way to discover and validate ideas, but it's a reference point.

because at the end of the day, it's the people who've actually used these products who really know.

If you've got a tool that tells you with 100% certainty that an idea is right — i.e. future profit data — please @ me like crazy, haha. Thanks again.

New on the RightIdea roadmap: pulling in G2 SaaS reviews too. Stay tuned.
I think a cross-validation approach combining the app and SaaS would be better, allowing both to be utilized. The main thing is that I intend to develop a new SaaS but have no plans to build an app right now; I don't own a Mac, and I have no desire to develop Android apps.

I'm really curious about how you did it—is there a better approach?