A serial app builder who's launched 14+ apps shares his complete process for generating SaaS ideas and a step-by-step validation framework for deciding which ones to actually build.
How to Generate App Ideas
- Just build anything (seriously): Building and launching an app exposes you to pain points in the growth process — those pain points become your next app ideas. It's an endless flywheel.
- Study social media ads: The algorithm targets you with apps for your demographic. If an ad reaches you, it's validated for people like you — and you already know how to reach that audience.
- Go on side quests: Meet people in other industries and discover their pain points. Example: Freed (AI medical note-taking) was built by a tech person whose partner was a physician — it's essentially a Cluey/Granola clone, but hyper-targeted to an underserved market.
- Recreate your favorite apps: If you're a paying power user of a tool, you already have a validated market, you're the target user, and you know others who'd use it too.
- Write everything down: Keep a running list of every idea (220+ and counting). 99.99% will be garbage, but the habit ensures you never lose a potential winner.
The Validation Framework
1. Can You Market It?
This is the first and most important question. If you can't market it, don't build it. Marketing channels to consider:
- Reddit marketing: Free, fast validation. Use tools like F5bot to monitor keywords. Con: subreddit mods can be aggressive about promotion.
- SEO: Free traffic that compounds over time. Con: takes months to build credibility — poor for quick validation.
- Paid ads: Unlimited growth engine when it works. Con: expect to invest thousands before finding what works.
- Organic social media: Low cost, fast validation on TikTok/Instagram. Con: video creation is a skill you need to develop.
2. Can You Technically Build It?
Less of a concern in the AI era, but still worth considering honestly.
3. Does It Have Existing Competitors?
"If a current idea you're trying to build truly has no other competitors, I think that's more of a red flag than a green flag."
Competitors = validated market with paying customers. Solo developers can't afford to create new categories — leave that for VC-funded teams. With competitors, there's zero market risk; it's purely product and execution risk, which you can control.
4. Can You Differentiate?
Not critical early on — you can copy an existing product and reach $1K MRR (his meeting transcription clone Monty hit $1,500-1,700 MRR as a straight copy). Traditional moats barely exist anymore thanks to AI. Differentiate through:
- Better design
- Better pricing
- Better customer support
- Time to value (TTV): The fastest path to a delightful experience wins
5. Are You a User of Your Own Product?
Building something you personally use keeps motivation high through the inevitable hard times, leads to better products through constant dogfooding, and makes marketing natural since you know the audience.
The bottom line: The best strategy combines ideation and validation in one step — recreate an app you already pay for. It has a validated market, you're the target user, and you know how to reach similar people.