Turn your idea into real evidence.[1]
Buteo is an AI idea validator. It researches your market, hunts for the reasons your idea fails, and checks every source. You get a Continue, Pivot or Kill call with the evidence behind it. The decision stays yours.
[1] Every finding cites a page you can open, and a claim that page does not support is flagged, never hidden.
Follow one idea from a paragraph to a verdict.
Built for founders who would rather hear “kill it” now than in a year. Read the full method
Example idea · illustrative, not a real run · a real run links every source
- 01
Your idea becomes claims
Every idea rests on assumptions: the problem is real, people will pay, nobody owns it yet. Each one is written down so it can be tested.
Your paragraph
A tool that sends polite payment reminders for freelance designers, so they stop chasing late invoices themselves.
Problem
Designers lose time to late invoices.
CriticalMarket
Enough freelance designers invoice their clients directly.
Competition
No tool is built for solo designers.
Feasibility
Reminders can be sent from the invoicing tools designers use.
Business model
Solo designers will pay for reminders.
CriticalRisk
Clients do not resent an automated reminder.
- 02
Five researchers, side by side
Up to five work at once: market, customers, competitors, money, risk. Two can never be switched off: the one hunting for competitors and the one hunting for reasons you fail.
5 research passes · side by side
- Market & web research
- Competitor researchAlways runs
- Customer research
- Business model research
- Risks & contradictionsAlways runs
- 03
Every source is checked
A finding only counts if the page it cites actually says it. When it does not, the finding is flagged in plain sight.
- Freelancer forum threadQuality 58%
Designers on a freelancing forum describe chasing overdue invoices as a recurring chore, and several say they have dropped a client over it.
For: Designers lose time to late invoices.Community or socialCustomer research - Invoicing tool product pageQuality 43%
An established invoicing tool already sends automatic payment reminders, and names freelancers among the people it is built for.
Against: No tool is built for solo designers.Vendor's own claimCompetitor research - Pricing survey write-upQuality 24%
Solo designers are willing to pay a monthly fee for automated payment reminders.
For: Solo designers will pay for reminders.Cited page does not say thisBusiness model research
- 04
We attack our own best evidence
Before the verdict, a last pass tries to break the strongest findings. What survives, stays.
Recommended
Continue
Late payment is a recurring pain for freelancers, and they already pay for tools around it. The space is crowded for agencies and thin for solo designers. Nobody has published what a solo designer will pay, so that is the first thing to test.
Why this callHide reasoning
Problem and market evidence is strong and comes from independent sources. Competition and pricing are mixed: the gap is real, the willingness to pay is unproven. Continue, and run the pricing test before building.
65of 100Evidence confidence
Mixed evidenceEvidence by dimension
ProblemMarketCompetitionFeasibilityBusiness modelRiskStrong evidenceMixed evidenceThin evidence
6 of 6 dimensions scored · The call is yours — nothing here closes an idea on its own.
What a single validation did.
- searches
- 55
- findings
- 52
- competitors mapped
- 8
- to a first verdict
- 13 min
Measured on one real run. Yours will differ.
From a problem space to a build plan.
- 01
No idea yet
Find an angle
Start from a problem space and get distinct angles worth testing.
- 02
You have an idea
Validate it
Claims, research, a checked ledger, and a Continue, Pivot or Kill call.
- 03
It held up
Take it to market
A cited go-to-market study: channels, demand, pricing, a 90-day plan.
- 04
Time to build
Plan the build
What goes in v1, the stack and the roadmap, drawn from the evidence your run found.
Start free. Then pay for what you run.
Buy a pack once, or subscribe for a monthly allowance at a lower rate. If a run fails, its credits come back.
Your first validation is free.
Confirm your email and validate one idea, start to verdict. That is 10 credits, on us. No card, and nothing renews.
10free credits · one validation
One-time
Builder pack
$49
100 credits
$4.90 per validation
Ten validations, or a mix with studies and build plans. Never expires.
Start freeMonthly
Pro
$19 / month
60 credits each month
$3.17 per validation
60 credits every month, at half the pack rate.
Start freeMonthlyBest rate
Scale
$49 / month
200 credits each month
$2.45 per validation
200 credits every month, for validating as a habit.
Start free
Every account starts on the free validation. Packs and plans are bought once you are in, from Settings → Billing.
What a credit buys
- A validation, start to verdict
- 10 credits
- Another research pass on the same idea
- 8 credits
- A go-to-market study
- 8 credits
- A build plan
- 6 credits
- An idea-generation session
- 1 credit
What founders ask before they start.
What is Buteo?
Buteo is an AI idea validator for founders. You describe a startup idea in a paragraph. Buteo turns it into testable claims, researches them on the open web, checks every source it cites, and gives you a Continue, Pivot or Kill recommendation with the evidence behind it.
How does Buteo validate a startup idea?
It writes a claim for each of 6 lenses: problem, market, competition, feasibility, financial, and risk. Up to five research agents then work in parallel, and the competitor and risk passes always run. Every finding is checked against the page it cites, and a final pass tries to break the strongest evidence before the verdict.
Does Buteo decide for me?
No. Buteo recommends and you decide. A run pauses at four points for a human: to clarify the idea, to approve validation missions, to report what real experiments found, and to choose Continue, Pivot or Kill. An idea is never killed automatically.
How long does a validation take?
Minutes, not seconds. One measured first pass took 13 minutes to reach a verdict, ran 55 searches, kept 52 findings and mapped 8 competitors. That is one real run of one idea, so yours will differ.
What does the confidence score mean?
Confidence is evidence strength, not idea quality. It says how well the sources Buteo found support each claim, lens by lens. There is no single viability score, because nothing in the analysis computes one. The verdict leads, and each confirmed point cites a page you can open.
What happens after the verdict?
You can test the open questions with real customers and re-evaluate as results come in. If the idea holds, Buteo writes a cited go-to-market study and a build plan drawn from the same evidence. If you have no idea yet, it can propose angles from a problem space.
What does Buteo cost?
Your first validation is free and needs no card, only a confirmed email address. After that Buteo runs on credits: a validation costs 10, a go-to-market study 8, a build plan 6 and an idea-generation session 1. A pack of 30 credits is $19 and never expires, and a monthly plan starts at $19 for 60 credits. If a run fails, its credits are returned.
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 and verdict, never your evidence, hypotheses or scores, and you can unshare it at any time. Go-to-market studies are different: a finished one joins a library every signed-in user can read, and may be featured on the public site with its author's agreement.
Bring the idea you are least sure about.
Describe it in a paragraph. Buteo does the reading, and shows you every page it read.