How to write a YC application that survives diligence (specific examples).

Y Combinator receives approximately 30,000 applications per batch and accepts roughly 1 percent. The math implies that most applications are filtered quickly, on a small number of signals. The applications that get interviews are dense with specificity: specific customers, specific revenue, specific problems, specific differentiation. The applications that get rejected are vague on most or all of those dimensions, regardless of how interesting the underlying idea is.

The YC application has a small number of high-leverage fields. The "what is your company going to make" field is the first one a partner reads. A vague version: "We are building an AI-powered platform that helps companies make better decisions about their data infrastructure." A specific version: "We sell automated data warehouse migration to mid-market companies running PostgreSQL who need to move to Snowflake. We charge $25K per migration. Our first three customers paid us $75K combined in the last sixty days."

The specific version answers four questions that the vague version does not. Who is the buyer (mid-market companies running PostgreSQL). What is the product (automated data warehouse migration to Snowflake). What is the price ($25K per migration). What is the traction (three customers, $75K, sixty days). The reader can evaluate the opportunity in twenty seconds. The vague version requires the reader to extract these answers from a one-on-one interview, which the founder will not get because the application did not earn it.

The traction question is where most rejected applications fail. Founders without revenue often leave it blank or write something like "we have not launched yet, but we have validated the problem through interviews." This answer is functionally a zero. YC partners read it as "no traction."

The fix when you have no revenue is to substitute the most credible non-revenue traction signal. A waitlist of 500 qualified email addresses is weak but better than nothing. Five paying design partners at $0 (free pilots) with signed letters of intent is stronger than the waitlist. A working prototype with 100 weekly active users is stronger still. The hierarchy YC partners apply, roughly in order of strength: paying customers > non-paying users actively engaged > letters of intent > qualified waitlist > unqualified waitlist > "we validated through interviews."

A founder applying without revenue should write specifically about whichever rung they have reached. "We have a working prototype with 87 weekly active users acquired through Hacker News. They use the product an average of 3.2 times per week. The most engaged user is a senior engineer at a Series B fintech company who has integrated our API into his team's deployment pipeline." That paragraph is a credible traction story even without dollars.

YC's question about why your team is uniquely positioned is the one where founders most often reach for credentials that do not apply. A common pattern is to claim domain expertise on the strength of a single internship or a college research project. YC partners read this as "no domain expertise." The fix is to describe the actual depth of your experience honestly, then explain how you compensate for the gaps.

A specific example: "Two of us spent four years inside Stripe's risk team working on the exact fraud problem we are solving. The third cofounder is a research engineer with three papers in graph neural networks, which is the technical approach our product uses. Our weakness is enterprise sales, and we are recruiting an advisor with ten years at Salesforce to address it." That paragraph tells the partner what the team is good at, what they are not, and what they are doing about it. It is more compelling than a paragraph that claims expertise across every dimension.

The application asks about competition. The wrong answer is "we have no direct competitors." Every YC partner reading that sentence interprets it as one of three things: the founder has not looked hard enough, the founder is in a market too small to support competitors, or the founder is in a market that does not exist. None of these is the answer the founder wanted to send.

The right answer names three to five competitors specifically, explains how each one is different from your approach, and identifies the segment where your approach wins. The structure to use: "We compete with [Company A] at the small business segment, where we win on price. We compete with [Company B] at the mid-market segment, where we win on integration depth. We do not compete with [Company C], which targets enterprise and uses a fundamentally different architecture. The gap we are filling is mid-market companies that need integration depth without enterprise pricing or implementation timelines."

YC requires a one-minute video of the founders. The video is not evaluated on production value. It is evaluated on whether the founders can articulate the business clearly under time pressure, whether they seem like the kind of people who can recruit and sell, and whether the team dynamic looks functional. Founders over-produce these videos in ways that hurt them. A grainy webcam recording of the team explaining the business clearly beats a polished video with slick graphics and weak content.

Applications that earn interviews are evaluated again on specificity in the interview itself. The interview is ten minutes, fast-paced, and partners interrupt frequently. The questions are not about the deck. They are about the business. "Why did you charge that customer $25K instead of $15K?" "What happens to your unit economics if your conversion rate is 30 percent instead of 60 percent?" "Who are the three companies you would compete with if you scaled tomorrow?" Founders who have specific answers to these questions on demand get in. Founders who pause, hedge, or generalize do not.

YC applications that earn interviews are dense with specific information: named customers, specific revenue, named competitors, honest gaps. Vague applications get rejected regardless of how interesting the underlying idea is, because the partners reading them cannot evaluate the opportunity in the time available. The work to make an application specific is the same work that makes you a better operator, so doing it well has compound value even if you do not get in. Most applications that earn interviews also rate well with investors who later see them, because the discipline of specificity is the same discipline that closes seed rounds. For the underlying analysis that strengthens both, see how to validate a startup idea and Verdikt's report methodology.