How to Make a Hiring Decision You Won't Regret

The Moment Where Hiring Goes Wrong

Most managers think the hard part of hiring is the interview. Finding the right questions, reading the person across the table, figuring out in 45 minutes whether someone is who they say they are.

That's hard, sure. But it's not where most hiring mistakes actually happen.

The mistake happens after. In the debrief, when five people sit around a table (or a Zoom call, if you're unlucky), and someone says "I just really liked them" and nobody pushes back. The room defaults to the loudest voice, or the most senior person, or whoever had the best story to tell about their interview. The hire gets made. And six months later you're having a completely different kind of conversation about that person.

I've conducted well over a thousand interviews as a Bar Raiser at Amazon, and the pattern I saw most consistently was not bad interviews. It was bad decisions made after good interviews. People who could run a rigorous process for 60 minutes and then abandon all of it in the final 10.

So let's talk about how to actually decide.

Start Before the Debrief

Here's something most teams skip entirely: written feedback before anyone talks.

The reason this matters is simple. The moment you hear a colleague's opinion, your own memory of the interview starts shifting toward theirs. It's not dishonesty, it's just how brains work. We are social animals and we update toward consensus automatically.

The fix is low-tech. After every interview, the interviewer writes down their assessment before they hear anyone else's. What did they observe? What did the candidate actually say? What did they see or not see? And then, the actual recommendation: hire or no hire, and why.

This doesn't have to be a five-paragraph essay. A few substantive sentences per topic area is enough. The goal is to lock in the independent signal before the group dynamics erase it.

If you do nothing else differently after reading this, do that.

Have a Rubric Before You Have a Candidate

The second place hiring goes sideways is when the scorecard gets built around the candidate instead of around the job.

What I mean is this: you interview someone impressive, and suddenly the things they're great at become the things you care about most. You convinced yourself that charisma matters more than technical depth because the candidate was charismatic. Or you decided domain experience was less important than raw intelligence because the candidate had intelligence but not experience.

The rubric has to exist before the first resume comes in. What are the three or four things that someone actually needs to succeed in this role? Not nice-to-haves. Not impressive-sounding things. The actual requirements. And then: how will you evaluate each one during the interview? What would a strong answer look like? What would a weak one look like?

When you have that written down ahead of time, you have something to hold the debrief accountable to. Instead of "I just really liked them," you're asking "did they demonstrate X or not?" That's a different conversation, and it produces better outcomes.

The Debrief Is a Decision Meeting, Not a Storytelling Session

This one is about facilitation, and it matters more than most managers realize.

A bad debrief starts with the hiring manager asking "so, what did everyone think?" and then the person who went first sets the entire frame. Everyone else either agrees or mildly qualifies. You end up with manufactured consensus that has nothing to do with what six people actually observed independently.

A good debrief collects verdicts before it collects stories. The facilitator goes around the room and asks for each person's recommendation, hire or no hire, before anyone explains their reasoning. Then you look at the spread. If it's five hires and a no-hire, your job is to understand that dissent. Not to overrule it. To understand it. Because sometimes the person who said no hire saw something everyone else missed.

That dissent is valuable. The goal is not to get to consensus. The goal is to get to the right answer, even if that takes longer.

The Reversibility Test

Here's a framework I've used for years, and it's not complicated.

Before you make any hire, ask yourself: how bad is it if I'm wrong?

Some roles are highly reversible. You hire a junior analyst, it doesn't work out, you have a hard conversation and you part ways. Painful, but not catastrophic. You can recover fast.

Some roles are not reversible at all. A senior leader hire that goes wrong doesn't just affect the person, it affects the team they manage, the projects they own, the culture of the organization. Those mistakes take 12 to 18 months to fully unwind, and some of the damage is permanent.

The standard of evidence you need should scale with the cost of being wrong. For a low-reversibility hire, you should want near-certainty, not confidence. If you're sitting in a debrief thinking "I think they'll probably be fine," that's not good enough for a VP role. For a reversible role with a clear ramp period and close management, it might be.

Most teams apply the same standard everywhere. That's a mistake.

When the Room Is Split

Split verdicts are actually your best learning opportunity, and most teams treat them as a problem to resolve rather than a signal to understand.

When you have a genuine split, the question is not "who do we believe?" It's "what did the people on each side actually observe?" Sometimes a split reveals that two interviewers were evaluating completely different things. Sometimes it reveals that one interviewer had a 20-minute conversation and the other had a 60-minute deep dive and of course they came to different conclusions.

Occasionally a split reveals real information about the candidate: they were strong in structured situations and weak in unstructured ones, or they showed well on technical depth and showed poorly on collaboration. That's not a reason to dismiss the candidate, but it is a reason to ask whether those weaknesses are dealbreakers for the specific role.

Point being: a split in the room is data. Use it.

The Last Question Before You Decide

After all of it, after the rubric check and the debrief and the reversibility test, I have one final question I ask before making a call.

What would have to be true for this hire to fail?

Write it down. Seriously. What are the conditions or behaviors or situations that would make this person not work out? If you can write a specific, credible answer, you now have two things: a reason to keep looking if the risks are too high, and a management plan if you decide to move forward. You know what to watch for. You know where to focus the onboarding. You know what conversations to have early.

If you can't write a credible failure scenario, one of two things is true. Either this is an exceptional hire and your due diligence is done. Or you haven't been honest with yourself about what you observed.

Either way, that question forces the answer.

Decide, Then Own It

Here's the thing nobody says out loud about hiring decisions. You will not always get it right. You can run every framework in this article, and you will still occasionally hire someone who doesn't work out, because people are complex and interviews are imperfect and organizations change.

The goal of a good hiring process is not a 100% success rate. It's a defensible decision-making process that improves your odds over time and makes your failures informative rather than just painful.

When it goes wrong, go back to your notes. What did you see that you explained away? What did the dissenter say that you didn't weigh heavily enough? What did your rubric tell you that you overrode with gut feeling?

That's how you get better at this. Not by beating yourself up, but by treating each outcome as data.

The candidate who impressed the room is not always the candidate who will do the job. Your job in the debrief is to figure out the difference.