Validate a startup idea with evidence you can open
Buteo's idea validation turns a paragraph about your startup idea into testable claims, researches each one on the open web and in official statistics, checks every source it cites, and gives you a Continue, Pivot or Kill recommendation with the evidence behind it. You decide what happens next.
By Buteo · Updated
01
Your paragraph becomes claims
problem
- Problem-solution fit
- Value proposition
market
- Target market clarity
- Timing
competition
- Competition level
- Market entry barriers
feasibility
- Initial feasibility
- MVP viability
financial
- Resource requirements
- Business model
risk
- Legal & platform risk
02
Agents research them in parallel
Market & web research
Competitor researchAlways on
Customer research
Business model research
Build & feasibility research
Reach & channel research
Open-source alternatives
Risks & contradictionsAlways on
Who tried this and failedAlways on
Official statistics & filingsAlways on
03
A verdict, and you decide
Continue
Nothing the idea depends on is refuted. Keep validating, and see how firmly it stands.
Pivot
The problem looks real, this answer to it does not. Change direction while changing is still cheap.
Kill
A claim the idea depends on is solidly contradicted, and the problem is not established. Stop before the build.
How firmly it stands
Validated · Unproven · Contested · Refuted
How it works
Describe the idea
A paragraph is enough. A few questions come first, where an answer would change the research.
Research runs in parallel
Up to 10 agents at once. 4 always run: competitors, risks, who already failed, and official statistics.
Evidence is checked, then attacked
Each finding is held to the page it cites. A last pass then tries to break the strongest support.
You read the verdict
Continue, Pivot or Kill, how firmly it stands, and the tests worth running next.
What you get
A recommendation that shows its standing
Continue, Pivot or Kill, printed beside validated, unproven, contested or refuted. A Continue never reads as a green light.
The claims your idea rests on
A hypothesis on each of 11 pillars, the critical ones marked, with what the evidence says for and against.
A ledger of checked evidence
Every finding with its source, its quality, and whether the page supports it. What failed the check stays visible.
A competitor map
Who is already there, each row backed by a page the search returned.
Figures as sources state them
Market size, buyers, growth and prices appear only when a checked source states them. No model computes one.
The tests worth running
Real-world experiments for what research could not settle, each with its pass mark set in advance.
What does idea validation do?
It finds out, before you build, whether the things your idea depends on are true. Buteo writes those things down as claims, one for each of 11 pillars: that the problem is real, that a buyer can be named, that the alternatives leave something unsolved, that it can be built, that the price covers the cost. Then it looks for evidence for each claim and against it.
The case against your idea is not optional. Research into competitors, into risks, into companies that already tried this and shut down, and into official statistics runs on every validation and cannot be switched off.
How long does it take, and what does it cost?
About 2 minutes, on the one first pass that was measured. Before its verdict that run made 117 searches, kept 81 findings and mapped 4 competitors. It is one real run of one idea, so yours will differ.
A validation costs 14 credits. Your first is free: 14 credits, no card. See the pricing.
Why trust the verdict?
Because you can check it. Every confirmed point in the verdict cites a page you can open, and each finding is held to the text of that page. How much a source weighs is a fixed formula, not a model's opinion, and there is no overall score to take on trust.
Kill is the hardest recommendation to earn: it needs a claim the idea depends on to be solidly contradicted and the problem itself not established. Short of that the recommendation is changed to Pivot or Continue, and the report says so. The whole method is public.
What can it not tell you?
Whether strangers will pay. Buteo reads what has been published, so it cannot see demand nobody has written about. Where research cannot settle a claim it says so and proposes a real-world test that can. Continue means keep validating, not start building.
What happens after the verdict?
Nothing is required. You can run the tests it proposes and report the results, run another research pass for 11 credits, change direction, or close the idea. When you go ahead, a build plan is drawn from the evidence this validation checked, and a go-to-market study researches the launch.
Where it sits on the way to a startup
01 Find the idea
02 Validate
- Idea validation
- Source checking
- Competitor map
03 Test
04 Build
05 Launch
06 Grow
- Find leadsSoon
- Marketing campaignsSoon
Questions people ask about this
Do I need a business plan to validate an idea?
No. A paragraph in your own words is enough. Buteo asks a few questions before it starts, where an answer would change what it researches, then works through to its verdict without waiting on you.
Does Buteo decide whether I should build?
No. Buteo recommends Continue, Pivot or Kill and shows the evidence behind it. The decision is yours, an idea is never killed automatically, and an answer Buteo cannot read changes nothing.
Is my idea private?
Yes, by default. A validation is visible only to you. If you choose to share an idea, other signed-in users see its title, summary, verdict and what you decided, never your evidence, hypotheses or scores, and you can unshare it at any time.