Most people look at G2 reviews to decide if a software tool is worth buying. Sales teams use them as social proof to close deals. But for founders and product builders, these reviews are something else entirely: a map of every gap in the market.
The real money isn't in building the next "all-in-one" platform. It’s in "unbundling"—finding the one specific feature customers actually use inside a bloated, expensive enterprise suite and turning that single feature into a standalone product.
Stop Looking for "New" Ideas
You don't need to reinvent the wheel; you just need to find a wheel that’s currently attached to a $50,000-a-year tractor.
Start by targeting categories dominated by massive platforms—think CRMs, project management giants, or HR software. These tools inevitably suffer from feature bloat. When a pricing page requires a "Contact Sales" button or includes a 15-page PDF to explain its tiers, it’s a prime target. You’re looking for the one tool users are desperate for, but currently have to pay a massive premium to access as part of a bundle.
Filtering the Signal from the Noise
Reading every review on G2 is a waste of time. To find an actual business idea, you have to ignore the extremes:
- Trash the 5-Star Reviews: These are almost always the result of a customer success team begging for a favor or offering a $20 Amazon gift card. "Great UI!" and "Friendly team!" are useless fluff.
- Ignore the 1-Star Rants: These are usually emotional reactions to a single bad support interaction or a billing dispute. They rarely tell you anything about the product’s functional flaws.
- Live in the 3 and 4-Star Zone: This is the sweet spot. These users like the software enough to keep paying for it, but they’re frustrated enough to write down exactly what’s broken. They are your most honest consultants.
Keywords to Watch For
When scanning these reviews, look for these specific red flags:
- "Overkill": This is a polite way of saying "I'm paying for 90% of stuff I don't use."
- "I only use...": This is the holy grail. If dozens of reviewers say they only use the reporting widget or the automated invoicing, you’ve just found your standalone product.
- "Steep learning curve": This means the current solution is too complex. If you can build a version that takes five minutes to set up instead of five weeks, you win.
- "Expensive for what it is": This confirms the market is ready for a leaner, cheaper alternative.
Scoring Your Discovery
Once you’ve found a recurring complaint about a specific feature, you need to see if it’s actually a viable business. Ask yourself:
- Frequency: Is this a one-off gripe, or is every third person saying it?
- Independence: Can this feature survive on its own? An "email signature generator" is a business. A "better delete button" is just a feature request.
- The Anger Factor: Are users mildly annoyed, or are they losing money? Anger is a much better sales driver than "mild inconvenience."
Moving from Research to Building
The biggest hurdle for most founders is the manual labor. Tagging and sorting 2,000 reviews by hand takes weeks, and most people give up before they find the pattern.
This is where automation helps. Instead of spending months on manual spreadsheets, tools like Feature2Product can scan thousands of reviews to pull out the "I only use X" sentiments automatically. It calculates a "productizability" score based on real user data so you don't have to guess.
The goal is to reverse the traditional startup path. Instead of building a prototype and praying someone wants it, you find exactly what people are already complaining about, verify the demand in the data, and build the solution they’re already asking for.