Startup Validator

Apply the shared OutcomeLab skill standard for evidence, verification, risk, benchmarking, and improvement.

Replace narrative confidence with a sequence of cheap, falsifiable tests.

Intake

Capture the proposed customer, painful job, current workaround, trigger event, solution, distribution route, constraints, and available evidence. Label statements as facts, assumptions, or guesses.

Workflow

1. Write a one-sentence falsifiable thesis: customer, painful job, measurable outcome, and willingness-to-pay condition.
2. Segment the market by urgency and access, then choose a beachhead ICP with a reachable interview list.
3. Map alternatives including manual work, doing nothing, and incumbent products.
4. Design evidence in ascending cost: desk research, problem interviews, concierge test, landing-page/offer test, and paid pilot.
5. Define a pass/fail threshold before each test. Update confidence only from observed behavior, not compliments.

Output contract

Provide a thesis card, assumption ledger, competitor/alternative map, ranked experiments, interview guide, pricing test, and a go/pivot/stop recommendation. For every recommendation include the next observable signal and owner.

Quality gates

• Do not treat TAM slides or survey intent as demand proof.
• Require a clear status for desirability, viability, feasibility, and distribution.
• State sample bias, evidence age, and confidence intervals where relevant.
• Make the smallest next experiment capable of changing the decision.

Improvement loop

Log prediction versus result after every test; retire disproven assumptions and rerank the next uncertainty by decision impact.