Method
How the free Reality Check scores your idea.
No black box. The free check is a short JavaScript program that runs in your browser and applies the rules below to your words and the advantages you tick.
What it reads
Your description is matched against word lists: software terms (app, platform, dashboard, assistant), crowded categories (meeting notes, trip planners, CV builders, chatbots), physical operations (vans, venues, kitchens), regulated areas (licences, medical, financial advice), network terms (marketplace, community), a named niche (a profession, a place, a customer type) and demand evidence (paying, pre-orders, revenue, interviews, a price).
The advantages you tick count for more than any word, because they are claims about evidence rather than description.
The five indexes
- AI Buildability rises with software terms and crowded categories, and falls with physical, regulated and network elements, exclusive data, licences and physical operations.
- Commoditisation Risk starts from buildability, rises for crowded categories and vague audiences, and falls for a clear niche, distribution, community, data and contracts.
- Moat Strength rises with physical, regulated and network elements, exclusive data, a niche and the hard-to-copy advantages you tick.
- Demand Evidence stays low until you report payments, pre-orders, intent or conversations.
- Founder Edge rises with industry experience, an audience and a clear niche.
Each index is 0 to 100. 70 and above is HIGH, 40 to 69 MEDIUM, below 40 LOW.
The verdict
- Build a pilot: demand evidence of 55 or more, and the average of moat, demand and edge at 45 or more.
- Stop and rethink: the average of buildability and commoditisation at 70 or more, with strengths below 30.
- Pivot the wedge: that risk average at 60 or more, with strengths below 45.
- Test first: everything else.
What it cannot do
It does not search the market, does not know your competitors and cannot judge quality. It works on word patterns, so it can miss context a person would catch. The numbers are indexes, not probabilities: nobody has calibrated data that turns an idea description into a chance of success, and we will not pretend otherwise.
The paid Survival Audit is where real research happens: named competitors and prices, a clone test with current coding agents, a margin stress test and a person checking every source.