Table of Contents
We put a lot of weight on interviews for something that isn’t actually very good at predicting performance.
The average unstructured interview has a validity coefficient of around 0.38. In simple terms, it explains roughly 14% of the variance in job performance.
Yet this is still how a lot of us hire.
We sit across from someone for an hour. We have a good conversation. They seem sharp. They communicate well. Maybe we even think, I could see myself working with this person.
Then six months later we’re wondering how we got it so wrong.
The problem isn’t necessarily our ability to read people.
The problem is the process we’re using to read them.
A structured interview process forces us to stop asking, “Do I like this person?” and start asking, “What evidence do I have that this person can actually do this job?”
That’s a very different question.
Why Unstructured Interviews Get Us in Trouble
An unstructured interview feels natural, which is probably why we like it.
We start with a few questions, something interesting comes up, we follow it, the conversation goes somewhere else, and an hour later we feel like we really got to know the candidate.
Maybe we did.
But now another candidate comes in and we have a completely different conversation.
So what are we comparing?
One candidate may spend 20 minutes talking about leadership. Another spends 20 minutes talking about technical experience. A third happens to share an interest with us and the conversation takes off from there.
We think we’re comparing candidates.
We’re actually comparing different evidence.
And because the process is loose, the candidate has a lot more ability to control the story. Someone who interviews extremely well can create a strong impression without giving us much evidence that they can produce the outcomes the role requires.
That’s where a structured interview process becomes valuable.
A Structured Interview Process Starts Before the Candidate Arrives
The interview shouldn’t begin with us deciding what we want to ask while we’re sitting across from the person.
The questions should already exist.
They should be tied to the specific competencies we’re trying to assess, and every candidate interviewing for that role should be asked the same questions in the same sequence.
Then we score the answers against the same scale.
That sounds simple, but think about how different that is from the way many founder-led businesses hire.
We’re removing some of our ability to improvise because the improvisation is part of the problem.
We’re trying to create a fair comparison.
The strongest questions are behavioural.
Instead of asking:
“What would you do if you had to deliver a project under significant time pressure?”
Ask:
“Tell me about a time when you had to deliver a project under significant time pressure. What did you do, and what was the outcome?”
One gives us a candidate’s theory about themselves.
The other gives us something they actually did.
Use Past Behaviour, Then Keep Digging
A good behavioural question is only the beginning.
We still have to figure out what actually happened.
The STAR framework gives us a simple way to do that. We want to understand the Situation, the Task the candidate was personally responsible for, the Actions they took and the Results they produced.
The word personally matters here.
Candidates naturally say things like, “We increased sales,” or, “We implemented a new process.”
Okay.
What did you do?
We aren’t trying to catch someone. We’re trying to separate the team’s accomplishment from the candidate’s contribution so we can decide whether that experience is actually relevant to the role we’re filling.
A structured interview process gives us a consistent way to do that across every candidate.
One Interview Shouldn’t Carry the Whole Hiring Decision
For a significant hire, one conversation is asking too much of the interview.
A stronger process can move through three or four stages.
The first is a 30-45 minute screening interview to establish whether the person meets the basic requirements.
Then comes a 60-90 minute competency interview focused on the four or five competencies that matter most for the role.
After that, we can test whether they can actually do the work.
For a finance position, that might mean financial modelling. For a commercial role, it could be a case study or pitch. For operations, it might be solving a process design problem.
The exercise should look as much as possible like the work we’re actually hiring them to do.
Finally, a senior leader can assess cultural and strategic alignment, including whether the candidate’s ambitions and behaviours fit where the business is trying to go.
Now we’re no longer asking one interview to tell us everything.
We’re collecting different kinds of evidence at different stages.
The Structured Interview Process Still Fails If the Founder Overrides It
This may be the hardest part for founders.
We can build the entire process and then destroy it at the very end because someone senior says, “I just really like this person.”
That’s exactly why scoring needs to happen independently.
At least two interviewers should score the candidate before they discuss them with each other.
Otherwise, the person with the most authority can influence everyone else’s opinion before they’ve committed to their own assessment.
This matters because affinity bias is real. We tend to prefer people who feel similar to us. First impressions can also create halo or horn effects, where one strong or weak impression colours everything we see afterward.
A structured interview process doesn’t pretend we’re suddenly going to become unbiased.
It creates some protection against our biases.
If one interviewer sees a serious problem and another thinks the candidate is fantastic, we shouldn’t settle the disagreement based on seniority.
Go back to the evidence.
What did the candidate actually say?
What did they actually do?
What did the exercise show?
Better Hiring Is Part of Building a Business That Can Run Without You
This is where hiring becomes bigger than HR.
If we’re trying to move from owner-operator to owner-investor, eventually we need other people who can produce outcomes without us standing over them.
That means the quality of the people we hire matters enormously.
And if our hiring process is mostly built around instinct, conversation and whether someone “feels right,” we’re putting a lot of faith in a method that doesn’t predict performance particularly well.
A structured interview process won’t make every hire work.
It does something more practical.
It makes us define what we’re looking for, collect comparable evidence, test whether someone can do the work and make the final decision based on something stronger than whoever had the loudest opinion in the room.
Because at some point, building a business that can run without us requires getting very good at deciding who we’re willing to trust with it.
And “I had a good feeling about them” is a pretty expensive way to find out we were wrong.
Justin D Maxwell provides family office and investment bank services to the lower midmarket to founders who want 8 or 9 figure net worths. You can learn more here: www.justindmaxwell.com or take our free assessment here: https://fielding.global/articles/diagnostic.html