Pipeline Recruiting

Pipeline, not searches

Brian Chesky describes how he thinks a CEO should approach talent: never run searches, always build a pipeline. "The best way is pipeline recruiting. You're constantly recruiting. You're constantly meeting people in advance of searches."¹ The pipeline runs on a simple rule in his telling: every meeting's job is to get the next meeting. Informationally meet the best people in a domain, ask each who the best two or three people they know are, and get introduced, so that when a role opens you already know the strongest candidates in the world for it.

Chesky contrasts this with ad-hoc hiring, where a role opens, a search firm returns fifty profiles, and a hire eventually emerges from a pre-filtered pool with self-selected, almost always positive references. Two failure modes are baked in, he argues: the choice is from a filtered pool rather than the best people, and the references are not real signal.

Start with results, work back to people

A recurring instruction in Chesky's account is to invert the conventional process. "Find an ad you like and figure out who made that ad. Start with results. Work back with people. Don't start with the resume."¹ Rather than screen credentials and then look at work, he traces a campaign, product, or piece of work he admires back to the person or team behind it.

The CEO owns the top of the pipeline

Chesky says he is co-hiring manager for the top 200 people at Airbnb, which he acknowledges is "very radical." His stated reason: "My executives should hire people so good they would never work for them without my help. If you can hire them without my help, we're not reaching far enough."¹ He describes spending two to three hours a day on recruiting and frames a trade-off, that every hour not spent recruiting is an hour spent managing, since better people need less management. He also suggests the first employee should probably be a recruiter rather than an engineer, because the recruiter unlocks every other hire.

Glyman's corroboration and one divergence

Eric Glyman describes the same doctrine from Ramp. He calls hiring mostly hunting done before the interview: going out to find companies and people exceptional at one skill set, asking them for advice, and asking who else is coming up.² He puts a metric on the founder's share, saying that done right, at least a third of his time goes to recruiting, and that Ramp's first board meeting was 80% reviewing LinkedIn profiles for every position.² Glyman dates Ramp's first recruiter to roughly the ninth or tenth hire and stresses the intent was a hunter, not an HR head. He also calls referrals the top hack, on the logic that "within one or two business days of working with someone you'll know more than in your entire interview process."³

One divergence is worth recording. Chesky has the CEO co-hire the top 200 externally, reaching for people so good they would not otherwise join his executives. Glyman's matched conviction at senior level runs the other way: he prefers to grow senior people from within rather than hire them externally. Both hold that the founder must own recruiting and run a pipeline rather than searches; they disagree on whether the senior bench is recruited or promoted. The founder-owns-talent posture and the habit of building the network continuously connect to mispriced-talent-pools and new-york-talent-arbitrage.

Tensions the founders note

At a five-person startup, spending half of one's time on recruiting is hard to sustain alongside building product; the advice fits companies past initial product-market fit better than the earliest stage. The CEO-as-co-hirer model also depends on significant infrastructure around the CEO. And self-referral networks can create monocultures, skewing the pipeline toward people already inside the ecosystem.