The Hiring Premortem

A hiring team finishes a debrief for a VP of Engineering search. Four panelists. Everyone liked the candidate. The hiring manager says "I think we're aligned" and reaches for the close. One interviewer shifts in her chair. She noticed something in the case exercise, a moment where the candidate did the work themselves instead of handing it to their team. She doesn't mention it. The group is settling, and raising a concern now feels like slowing down something that wants to happen.

The offer goes out. Six months later the VP has written half the architecture docs themselves, the direct reports are waiting for approval on decisions they should own, and the team's output looks exactly like it did on day one. The interviewer who shifted in her chair thinks, "I saw that".

When the frame is "should we hire this person," speaking up feels like slowing things down. The group has momentum. The hiring manager is ready to move. Raising a concern means being the person who made everyone stay late. So people swallow their doubts, and the information that would have changed the decision stays locked inside individual heads. A premortem inverts that frame.

The technique comes from Gary Klein. I first encountered it in Kahneman's book Thinking, Fast and Slow, where he calls it one of the few bias interventions that actually works in groups. The idea is simple. You tell the group the decision has already failed, then ask them to explain why. When failure is assumed, raising a concern is just part of the exercise. The interviewer who noticed the delegation problem doesn't need to be the person who slows things down. She's doing what the exercise asks.

Klein's research suggests premortems surface 30 to 50 percent more failure modes than standard risk discussion, and more specific ones. Asking "do you have concerns?" invites someone to play the dissenter. Asking "write the story of how this failed before it failed" makes skepticism the task.

Say something like this:

"Before we move forward, I want to try something. It'll take 30 minutes. Imagine it's 12 months from now and this hire was a disaster. They're underperforming, or they've quit, or we had to let them go. I want each of you to write down what went wrong. Write as if it already happened."

Give everyone a piece of paper or ask them to open a blank doc. Writing prevents the loudest voice from anchoring the group before anyone else has formed a thought.

Watch for: Eye contact between panelists, looking to the hiring manager for permission. If people hesitate, add something like "This is hypothetical. We're stress-testing, not voting."

Everyone writes independently. No discussion or sharing yet.

This will feel unnatural for groups that talk through everything. Protect the silence. If someone starts speaking, redirect. "Hold that thought. Get it on paper first."

Once someone shares aloud, people start adjusting their answers to match what's already been said.

Watch for: People finishing quickly and looking up. Early finishers often wrote something safe and stopped. If most people are done in 90 seconds, say something like "If your first scenario was about skills or experience, write a second one about something else. Motivation, team dynamics, something about us, not just about them."

Each person reads one failure narrative aloud. The others should not respond, argue or add anything. Write each one on a whiteboard or shared doc where everyone can see it.

The first round will be polite and abstract. "Culture fit issues." "Didn't have the depth we needed." "Found the role wasn't what they expected." These are the acceptable answers, the ones that don't point at anything uncomfortable.

After the first round, push. "Those are useful. Now I want specifics. You mentioned they didn't have depth. What does that look like in month six? What meeting are you in when you realize it?"

That push changes the answers. Someone might say "Their references all came from the same company, and the answers about transitioning between roles felt rehearsed. I think six months in we find out the technical depth is narrower than it looked, and the team figures it out before we do." Or "Every example they gave about scaling was about hiring, none about making people work together. By month four the team is twice the size and half as productive because nobody is connecting the work."

If the hiring manager is generating the least specific responses, ask them directly. "What would make you regret this decision in a year?" Hiring managers who are invested in a candidate produce the weakest premortem material unless specifically prompted to imagine personal regret, not abstract failure.

In a panel of four, expect two people to produce 80 percent of the useful output. Your job is to get the two talking.

Watch for: Consensus disguised as premortem. "I think this person would fail if we don't give them enough support." That means "I want to hire them but I'm nervous." Push back. "Forget about what we could do to help. In your story, what goes wrong with the person?"

Group the failure narratives into themes on the board. Usually three or four emerge. Then ask the group, "Which of these would we see signs of in the first 90 days?" That tells you which concerns you can test with early milestones and which ones (like "they want the title, not the work") you'll just have to watch for.

The premortem does not make the decision. The group does. Sometimes the concerns are addressable, and the team builds them into the onboarding plan. Sometimes they reveal a gap worth testing with an additional reference check or assessment. Occasionally they surface a genuine deal-breaker that nobody was going to say out loud. Rare, but this is the entire point.

Don't force the group to a vote. Let the hiring manager make the call with the failure narratives on the table rather than buried in individual heads.

"We hired her to lead the post-acquisition integration, but her experience was all in organic team-building. By month four the acquired team had lost confidence in her because she kept building processes from scratch instead of adapting what they already had."

"He presented three successful product launches, all in the same market vertical. On the industrial portfolio he didn't have the customer relationships or the technical intuition. He spent the first six months learning what his peers already knew."

"She said she left her previous role because of leadership changes. Two of us independently wrote that we thought the real reason was performance management, and we hadn't checked that reference."

Now compare. "The candidate wasn't a good fit for the team." That sentence tells you nothing. If your premortem is producing output like that, the prompting in Step 3 needs more pressure.

The premortem becomes a confirmation ritual. When the hiring manager has already decided, the group reads the room and produces failure narratives that are easily dismissed. "They might struggle without enough support" means the group is being polite, not honest. The premortem only works when the decision is genuinely open. If you're running it after the hiring manager has already told the candidate "we're putting the offer together," you're too late.

The group stays abstract. If the failure narratives sound like "skills gap" or "cultural mismatch" after the push described in Step 3, the prompting wasn't specific enough. Go back and ask for specifics.

You run it for a role that doesn't warrant it. A premortem adds 30 minutes and works because the group has enough information to imagine specific failures. For a junior hire where the interview data is thin and the role well-defined, the premortem produces speculative fiction rather than risk analysis. Save it for senior and leadership hires where a bad call costs you six months, not three weeks.

Somewhere in every debrief, someone shifts in their chair and says nothing. The premortem gives that person a reason to talk. Thirty minutes are cheap against six months of watching a failure someone in the room saw coming.

The Recruiting Lattice takes mental models from fields like behavioral science, sociology, and decision theory and turns them into practical tools for talent acquisition.