You’ve done the work. You wrote the MOC. You defined competency layers and integrated your values. You did calibration calls to learn what great looks like. You know exactly what you’re hiring for. See: The MOC The document your entire interview loop is built from. Every question should trace back to it.

Now you need to evaluate candidates against that standard, and this is where most founders fall apart. They wing the interviews, and the people they hire wing it too. Nobody’s working from the same rubric, and no one compares notes until someone says “I think she’s great” and someone else says “I don’t know, something felt off.” That’s not a hiring process. That’s a group of people guessing separately and then arguing about their guesses.

Every team puts their hand on the hot stove at least once. The CEO meets someone impressive at a conference. Comes back to the office excited: “I think we could get this person.” There’s no open role. No MOC. No interview loop designed. The CEO brings them in anyway. Nobody knows what they’re evaluating. The interviews are just conversations. The person joins. Six months later they’re a bad fit for a role that was never defined, evaluated against standards that were never agreed on, with expectations that were never set. The financial cost is enormous. The social cost is worse: the CEO and the leadership team lose credibility. They had a process. They just didn’t use it.

Every piece of structure in this pack exists to stop exactly that.

Four gates stand between a candidate and an offer: screening, top-grading, focused interviews, and references. Before the how, the why.

The research isn’t ambiguous

A 2022 study in the Journal of Applied Psychology reanalyzed decades of hiring research. Structured interviews came out on top: the single strongest predictor of whether someone will actually be good at the job.¹ Not cognitive ability tests. Not years of experience, which turned out to be one of the worst predictors. Structured interviews. ¹ Revisiting Meta-Analytic Estimates of Validity in Personnel Selection Sackett et al. (2022) proved structured interviews are the strongest predictor of job performance, dethroning cognitive ability tests.

Google found the same thing. They compared tens of thousands of interview scores to on-the-job performance. Their famous brainteaser questions predicted nothing. Laszlo Bock, their head of people operations, was blunt: brainteasers “serve primarily to make the interviewer feel smart.” Google switched to structured behavioral interviews with consistent rubrics. A separate study found that narcissism and sadism predicted who uses brainteasers in interviews.³³ Dark Motives and Elective Use of Brainteaser Interview Questions The study that linked narcissism and sadism to brainteaser use in interviews.

You don’t have to take it from me. The research and the practitioners all land in the same place. Structure your interviews or accept that you’re leaving your most expensive decisions to chance.

Why founders skip this

Three reasons, and none of them are good enough.

First, they trust their gut. Founders who obsess over every pixel in their product, who wouldn’t ship a feature without user testing, who have a near-spiritual conviction about craft and quality, will wing one of the most expensive decisions they’ll make this year. They’ll spend months perfecting a button color and then hire a VP off vibes. They figure: I know what I’m looking for, I’ll know it when I see it. Your gut is not calibrated for this.

This is especially dangerous for one-off hires. Founders are more likely to build a structured process when they’re hiring multiples of the same role. But when it’s one VP or one director, they skip the process. Those are exactly the roles where you can’t slip, can’t cheat, can’t take the easy route. A bad account executive costs you a quarter. A bad VP costs you a year and the trust of everyone who reports to them.

Second, they’ve never seen it done well. If every interview you’ve been through was two people in a room making conversation, that’s your mental model. You don’t know what a structured process looks like because nobody showed you one.

Third, there’s a coordination problem. It’s one thing for you to decide to run structured interviews. It’s another to get five or six people on your team to do it consistently. That requires training, documentation, and someone who owns the process. Founders tend to see this coordination cost and think it’s bureaucracy. It’s not. It’s the cost of making a million-dollar decision with more than one person’s opinion. If you want to go invent a whole new approach and learn through painful trial and error, be my guest, but you’re going to be massively distracted from building your product. Hiring is one of the places where you shouldn’t fuck with it. This works.

Judgment is the point

Here’s the real reason structured interviews feel uncomfortable.

The purpose of a hiring process is to filter for high-performing people who can achieve your business objectives. Filtering people out requires making judgments, and that’s uncomfortable. Judgment is cast as a pejorative. You have to look at a person and decide: does this person fit the MOC, fit our values, or don’t they? And then you have to sort them in or out.

Instead of sitting with that discomfort, people are too nice. They’re more concerned with how candidates feel than with whether those candidates can do the job.

This is where the biggest pushback on the WHO method comes from. The book’s top-grading interview includes a move where you tell the candidate you’ll be contacting their former bosses by name. A lot of people, especially in HR, find this coercive. I think the specific “we will be calling your boss” framing is too heavy-handed. But the instinct behind the objection is worth examining, because there’s a dynamic where HR sometimes takes on a union-boss posture, feeling that their job is to represent and defend the interests and feelings of candidates. They don’t say this explicitly, but that’s the energy. The result is a process built for comfort, not signal.

You want employees to have a phenomenal experience at your company. That’s real, and it connects to how you think about employment as a product. See: The Other Product The people you build with are a product offering too. But in order to get great people in, you have to filter people out. Rigor is evaluating people against a standard. Cruelty is evaluating people against your ego.

The four gates

Every interview question should trace back to a specific competency, outcome, or value in your MOC. Every interviewer should know which piece of the MOC they’re responsible for evaluating. The interview loop draws on three other cards: Competency Layers (the three layers that go into every rubric), Values (your company-wide competencies), and The Calibration Call (how to learn what great looks like before you design the loop).

A complete interview loop has four gates. Each one filters for something different, and a candidate has to clear each one before they move to the next. This structure comes from WHO: The A Method for Hiring, and I’ve taught it to enough founders to know it works.²² WHO: The A Method for Hiring The hiring methodology this structure is built on. Smart and Street’s four-gate model, extended for how I teach it.

1. Screening eliminates candidates who aren’t a fit before you invest serious time. Thirty minutes on the phone. Four questions: what are your career goals, what are you great at professionally, what are you not good at or not interested in doing, and how will your last five bosses rate you when we talk to them? Only 10-20% of candidates should pass screening. This gate is roughly the same across roles.
2. Top-grading reveals the patterns in a candidate’s career. It’s a chronological walkthrough: for each prior role, what were you hired to do, what are you proud of, what were the low points, who did you work with, and why did you leave? You’re looking for whether they were pulled into roles (recruited, promoted) or pushed out. Whether their accomplishments tie to the expectations of the job or float free of results. Whether they own their low points or blame everyone around them. Smart and Street’s WHO has an excellent set of red flags to listen for during top-grading; it’s worth reading their chapter on the Select step before you run your first one.² This takes about three hours and uses the same structure for every role.
3. Focused interviews test whether the candidate can do the specific job described in the MOC. Each candidate does two to three, each about an hour, each digging into roughly two specific competencies or outcomes from the MOC. One interviewer might focus on strategic thinking and cross-functional influence. Another might test for role-specific technical skills. A third might evaluate values alignment. Screening and top-grading are consistent across roles. Focused interviews are where the MOC makes each loop unique.
4. References check what you’ve learned in interviews against the experience of people who’ve worked with the candidate. Seven reference calls total: the hiring manager does about four, other team members do the rest. For senior hires, talk to people who reported to the candidate, not just their bosses. The best reference question I’ve seen: “The candidate mentioned they struggled with _____ in that role. Can you tell me more about that?” You’re using the candidate’s own self-reported weaknesses as a starting point, and you’re signaling that you’ve done your homework.

Those are the four gates. Two things tie them together: the rubric and the debrief.

Build a rubric, not a vibe check

Before anyone interviews a candidate, build a rubric from the MOC. A rubric defines what excellent, good, fair, and poor performance looks like for each competency you’re testing. It turns “I liked her” into “she demonstrated strategic thinking at a level consistent with what we need for this role.” It keeps interviewers honest and gives you something to compare across candidates.

Score independently, then talk

Every interviewer scores the candidate against the rubric independently and submits their written feedback before anyone shares impressions. This is critical. If your interviewers share their views before they’ve written them down, you don’t have five independent data points. You have one opinion and four people, whether they realize it or not, adjusting to it. Exposure to other interviewers’ feedback before you’ve submitted your own warps your assessment. The effect is worst when junior people see senior opinions first.

After all interviews are complete and all feedback is submitted, the full team meets. The hiring manager circulates the written feedback in advance. A moderator leads the discussion. Some companies start with the most junior interviewer to avoid authority bias. The point is to surface patterns that no single interviewer would catch: a behavior that seemed minor in one interview looks different when three people noticed it independently.

Healthy disagreement is the goal. If everyone agrees immediately, worry about groupthink, not celebrate alignment. Probe the uncertainties. Ask where people disagree and why. There’s no such thing as a neutral opinion. If an interviewer isn’t convinced, that’s a no. By the end, there should be a clear recommendation, and the hiring manager makes the final call. But the call is only as good as the interviewers who fed it.

Train your people

None of this works if your interviewers aren’t trained. Effective interviewing is a learned skill, and it’s ridiculous to expect people to be good at it with no preparation. Even experienced people who’ve interviewed elsewhere should go through your process before they interview for you.

Get it on paper. Put together a deck or a document. Appoint someone to own the training. It doesn’t have to be your head of people; it can be the CEO, a co-founder, or just the person on your team who’s best at hiring. Everyone who’s going to be in an interview loop goes through an hour of training. You cover: why structured interviews matter, how the rubric works, what their role is in the loop, how to write feedback, how the debrief works, and what they legally can’t ask.

This isn’t something you do to be a pain in the ass. This isn’t something you do to bring bureaucracy. This is something you do to bring great people into the company. And great candidates love a disciplined process. They know that’s how you build a team that wins. Walk into any company you admire and they’ll have a serious hiring process. The fact that they have great people isn’t an accident.

There’s no easy button

The default in startups is chaos, not order. Your job as a founder is to bring the order. There’s no easy button for hiring. You build a great process, you train your team to run it, and you stick to it, because if you don’t, you fall prey to every bias that makes humans bad at this: wanting to move fast, falling for charisma over competence, convincing yourself that a bird in the hand is better than holding out for the right person. The process is what protects you from yourself.