Welcome to the show everybody! There's one hiring trend that almost everyone agrees is progress: skills based hiring. Stop screening on degrees and job titles, start testing what people can actually do. In NACE's Job Outlook 2026 survey, seventy percent of employers said they use skills based hiring for new graduates, up from sixty five percent the year before. So, good news?
It is good news, and I want to say that clearly before I complicate it. Testing what someone can actually do is far better than assuming it from a diploma or a famous company name on an application. But a skills test answers one question. Can this person do the work? It was never designed to answer the second question, which usually matters more. Will they?
Before we get to that, I have to push back on the good news a little. Harvard Business School and the Burning Glass Institute looked at whether companies that announced skills based hiring actually changed who they hired. Their 2024 report found that in 2023, the changes affected fewer than one in seven hundred hires. Lots of announcements. Very little change in practice.
That's an important correction, and it's a pattern we see with every hiring trend. The press release moves faster than the process. But let's assume a company does it properly. It removes the degree filter, it builds a solid work sample, and it tests every candidate fairly. It's still only seeing part of the picture.
Which part is it missing?
Think about the conditions of a skills test. The candidate is prepared, motivated, rested, and focused on one task for an hour. That tells you what they can do on their best day. The job isn't their best day. The job is month nine, when the client is difficult, the deadline moved, the manager is distracted, and nobody is watching. What someone can do and what they will do under that kind of pressure are two different measurements.
There’s data on where new hires actually fail. In an older but widely cited study, Mark Murphy and Leadership IQ surveyed more than five thousand hiring managers across three hundred twelve organizations, who together had hired more than twenty thousand people. Forty six percent of those new hires were considered failures within eighteen months. And when the managers explained why, eighty nine percent of the failures came down to attitude. Only eleven percent were about technical skill.
Eleven percent. So a well designed skills test is guarding the door that only one in ten failures walks through.
And the other reasons are telling. The biggest was coachability, at twenty six percent. Then emotional intelligence, then motivation, then temperament. I should be fair, though. That's hiring managers' judgment of why people failed, not a controlled experiment.
That's a fair caution, and worth saying. But the direction matches what I've seen for decades. People are rarely let go because they couldn't learn the software. They're let go because of how they operate when things get hard. Whether they take feedback. Whether they hold up under pressure. Whether they push or wait.
So what actually predicts performance, if skills only tell part of the story?
No single tool does. That's the first thing to accept.
That fits the most recent research. Paul Sackett and his colleagues re ran the big meta analyses on hiring methods in the Journal of Applied Psychology in 2022. They concluded that the validity of many traditional selection tools had been overstated for years. And the method that came out on top was the structured interview, with an average validity of about point four two. Good, but nowhere near perfect.
Which tells you the answer is layering, not replacing. Use the skills test for can. Use a structured interview, with the same questions and a scoring guide, for how they think. And add a measure of will: how someone is wired to operate under the specific pressure of that role. That last layer is the one most hiring processes simply don't have.
And that's where behavioral data comes in.
It is. A validated behavioral assessment looks at how someone tends to operate under the kind of pressure the role brings. And the most useful benchmark is your own top performers in that seat, so you’re not asking whether someone is impressive in general. You’re asking whether they operate the way your best people do. But it doesn’t replace your skills tests or your interviews. It answers the one question they were never built to answer. And it’s one structured input, never the sole basis for a hire.
So if someone listening wants to start this quarter, what's the first move?
Keep it small. Pick one critical role, the one where a mis hire hurts most. For your finalists, pair the skills test with a behavioral diagnostic. Then look back at the people you've already hired into that role. Who succeeded, who struggled, and what did the skills tests say about each of them? If the skills scores didn't separate your stars from your strugglers, you'll know exactly which question your process has been leaving unanswered.
And a good place to start is with the people who already succeed in that role. OAD's free Top Performer Profile takes seven minutes, results come back in twenty four hours, and there's no credit card. That's O-A-D dot A-I.
Anyone can perform on test day. Hire for month nine.
Perfect place to leave it. Thanks for the wisdom.
Always a pleasure.