Read your market
like an analyst.
In one afternoon.
Insight Engine is a structured research instrument for SaaS founders. Drop in an idea — get a deep, sourced map of the market, the competitors, the unit economics, and the signals that decide whether it’s worth building.
> problem_statement: "Too many APIs to manage"
> solution: "Unified dashboard for API keys"
> running_analysis
CAC > LTV in adjacent benchmarks.
From idea to verdict in three passes.
- 01
Signal collection
Our system collects evidence across the open web — Reddit, G2 reviews, competitor pricing, SEC filings, niche forums — and builds a structured map of the landscape with linked sources where available.
5–25 min · multi-source scan - 02
Risk synthesis
Structured analysis passes vet the findings. Failure modes. Founder fit. Action plan. Each pass can only lower the market score — never raise it. Honest by construction.
Honest by default · ~30 sec - 03
Verdict + plan
A combined score (0–10) and one of three calls: pursue, pivot, pass. Plus the single biggest strength, the single biggest risk, and concrete kill-conditions to watch for.
Structured report
Three things generic AI tools won’t do for you.
Most idea-validation tools are cheerleaders. Insight Engine is built to surface the strongest risks, tradeoffs, and reasons an idea may fail.
Real research, not recall
Our system collects evidence from the open web — pricing pages, forums, SEC filings, review sites — and produces a linked report designed for verification, not blind trust.
Honest by default
Findings are vetted by structured passes that can only lower the market score — never raise it. Optimism bias is treated as a bug, not a feature.
Tailored to you
Your founder profile — experience, budget, audience, time, risk tolerance — flows into every section. The same idea gets a different verdict for a 40h/week solo bootstrapper than a funded team of three.
The output is a decision aid, not a pep talk.
Every report culminates in a clear verdict preview, with evidence, risks, and a next-step recommendation.
Verdict
Strong founder-market fit based on prior open-source contributions. The core mechanic is well-understood, and technical execution risk is minimal.
Willingness to pay in the “personal productivity” space is notoriously low. Competitors offer similar features for free. Distribution will require massive organic reach.
Built for founders who’d rather know in 20 minutes than discover in 6 months.
You have 3 ideas. You can build one this quarter.
Run all three through Insight Engine in one afternoon. Pick the one whose verdict isn’t “pass.” Save 11 weeks of building the wrong thing.
You’re about to commit a designer + 2 devs for 6 months.
20 minutes of structured analysis versus 6 months of payroll. The tradeoff is hard to ignore once you compare the cost of checking first.
Pay per analysis.
New accounts get enough credits to try a Standard run first. If the idea is promising, buy more credits and go deeper.
Start with the included Standard run before spending on deeper research.
Common questions.
How is this different from asking a generic AI tool?
Generic chatbots give you a plausible essay drawn from old recall. Insight Engine collects evidence from the open web in real time, links sources in the report where available, runs structured passes that can only lower the market score, and tailors the analysis to your founder profile. Different category of output.
How long does a report take?
10–25 minutes from submission to verdict. You can keep working — analyses run in the background and the page updates when the report is ready.
Is my idea private?
Your idea text and the resulting report are stored on your account. They are not used to train any model.
Can the report be wrong?
Yes. The system can miss niche competitors, misread foreign-language sources, or quote outdated pricing. Treat the verdict as a structured second opinion, not as truth. The kill-conditions and action plan are designed to be tested against reality, not trusted blindly.
Why so many sections?
Because each section tests a different failure mode — market size, unit economics, distribution fit, defensibility, founder fit, and more. The exact structure can evolve, but the point stays the same: force the idea through multiple lenses before you build.