The framing, and it is an engineering one: idea lists are calibrated for a reader they never name — funded, full time, more than one pair of hands. Prose can leave that implicit. A scoring system cannot, and having to encode it is what forced the precision below.
The decision: exactly one class of check is resource-sensitive — whether the founder can clear a regulatory or compliance cost of entry. Every other check scores identically no matter what is in the bank.
The predicate is a conjunction on purpose: meaningful capital AND full time AND more than one person. Miss any one and nothing changes. A well-capitalised solo founder working nights does not qualify, because clearing payment rails and enterprise-trust compliance is a coordination job as much as a cheque.
The clearable list is short and specific: payment-card and information-security certification, money transmission, know-your-customer and anti-money-laundering, payment processing licences and gateways, an acquiring bank, card network membership. What they share is that each is a known process with a price and a timeline. Nobody has to invent anything.
The prohibited list is deliberately much broader: medical and health, diagnosis and therapy, prescription and controlled substances, gambling, securities and regulated financial advice, lending and insurance, the practice of law, anything involving minors, weapons, genetic testing, elder care.
The asymmetry between those two lists is the whole design, and it runs one way. An over-broad prohibition keeps an idea killed, which is recoverable — the user argues with it and we look strict. An over-broad allowance quietly un-kills a safety-critical idea, and nobody arrives to argue. Size the error you can afford.
The check that does NOT move with money, and this is the one that makes the feature honest: capital intensity. An idea that needs to buy inventory, build physical infrastructure or subsidise usage until scale arrives fails identically for a funded team and for one person. Making it resource-sensitive would wave through exactly the capital-heavy ideas that have historically consumed enormous funding and produced nothing.
The ratio worth stealing if you build anything that scores per-person: of the 12 deal-breaker checks a candidate runs through, exactly one responds to the founder's balance sheet. The temptation is to make far more of them situational than is true, and every one you add makes the system agree with whoever has the most resources.
Full write-up, with both lists in full: https://whittleos.com/guides/y-combinator-startup-ideas
Top comments (0)