How I Generate +100 SaaS Ideas

Generating SaaS ideas at scale isn’t about finding a single “genius” concept—it’s about building a repeatable pipeline that turns public customer frustration into testable product bets. The core approach centers on starting with established markets, mining complaints for specific pain points, and then validating the pattern before committing time or money. That matters because most ideas won’t pan out; the goal is to run enough structured experiments that at least one can become profitable.

The framework begins with choosing a market where demand already exists. Instead of inventing from scratch, it targets “busy internet streets” with heavy traffic: users already know the category, and the opportunity is to serve people who are unhappy with the current tools. A practical method is to use Reddit to search for “alternative to [tool]” and open posts that look relevant. Each candidate is then filtered using two questions: whether building an alternative is realistically viable, and whether the complaints can actually be addressed. The process intentionally weeds out ideas that are too ambitious, too uncertain, or based on issues that may not be solvable.

Once a shortlist forms, the next step adds stronger evidence than a single Reddit comment. G2 is used to locate product reviews for the tool being targeted, with a focus on the lowest-rated feedback. The goal is to find recent, specific complaints—ideally reviews from the last three months—that describe concrete pain points and what users would have wanted instead. The method calls for gathering at least five to ten additional reviews and then looking for common themes. If multiple users independently mention similar issues—like laggy performance, difficult navigation, or missing features—those repeated signals become the raw material for a new SaaS concept. In the example, the pattern points toward a “more lightweight” alternative that’s easier to use and runs faster in the browser.

After qualitative research, the process adds a newer, higher-volume tactic: collecting visual evidence of bad reviews and using ChatGPT to synthesize them into a business case. The workflow suggests using a Chrome extension (Go full page) to capture multiple full pages of negative reviews from sites like G2, then optionally expanding to sources such as Trustpilot or Facebook. The images are uploaded to ChatGPT, which produces a detailed business case and a specific outline for a SaaS idea. The emphasis here is less on perfect data quality and more on generating enough structured insight to spot patterns.

Before building, the framework introduces a cooling-off period. The idea should be added to an idea database, then revisited after one or two weeks to see whether excitement fades or persists. Persistent interest signals a stronger candidate. Even then, the advice is to stay diversified: push the best idea forward alongside other ideas rather than betting everything on a single concept. The end goal is an ongoing “SaaS factory” that continuously generates and tests multiple bets, accepting that only a few will succeed.