The 'Compounding Referrals' Tactic
Successful founders are treating every discovery call as a referral engine. The tactic, highlighted in a new guide, is simple: end every conversation by asking, "Who else do you know who is struggling with this?" to build a self-sustaining pipeline.
YC Partner Michael Seibel advises that a startup's first 10 customers should come from their personal network or from people willing to work with an early-stage, imperfect product. The key is to find "qualified customers" who feel the problem so intensely they are willing to pay for a solution, as this validates the urgency and value. Before a product exists, founders can find early adopters in niche online communities like Reddit, Discord, or industry-specific Slack groups. The strategy is to participate authentically for weeks by adding value and answering questions before ever mentioning your product, transforming you from a seller into a trusted community member. For cold outreach, YC Partner Gustaf Alströmer recommends plain text emails of no more than six to eight sentences. The message should clearly address the customer's problem, state that you're the founder, explain why your team is impressive, and include social proof. The goal of the first interaction isn't to sell, but to spark curiosity and build credibility. Instead of a Minimum Viable Product, YC General Partner Ankit Gupta suggests building a "Minimum Evolvable Product"—something simple that can adapt based on feedback from the first few users. These early adopters, who are often not price-sensitive, provide sharper feedback when they pay, helping to shape the product's direction. Udemy co-founder Gagan Biyani struggled to get his first course instructors, so he created his own course as a case study. After generating $30,000 in a few weeks, he used that success story as undeniable proof to attract the next wave of users, creating a scalable process for finding leads. The objective of these initial conversations is not to validate an idea but to understand a customer's problems and workflow. YC Partner Gustaf Alströmer advises asking open-ended questions like "What's the hardest thing about X?" and "How are you solving this today?" to uncover the underlying needs.
