I have interviewed thousands of people since 2006. My honest conclusion is that an interview gives you around 5–10% of the information you need to make a good hire. Yet for most CEOs and senior leaders, it is still the main selection tool.
That is not because interviews are useless. It is because we ask them to carry a weight they cannot bear. A conversation rewards people who are good at conversations, and in a senior leadership hire, that is almost everyone on the shortlist. Over the years I have refined hiring processes with many clients, and the approach below consistently works better than another round of panels.
Make the first interview shorter, and turn it around
Keep it to 30 minutes. Bring a colleague, ideally someone very different from you in personality, to limit the halo effect that one confident first impression can create.
Then don’t open with the company pitch. Give no background at all. Simply ask the candidate what questions they have.
This puts them in control and at ease, and it is revealing. The questions they ask, and the order they ask them in, tell you what genuinely matters to them. It also protects you from confirmation bias: while they are asking, you are not busy asking questions designed to confirm the view you formed in the first two minutes.
When they have finished, ask one question only: “What do you consider to be your proudest work achievement so far?” Then probe for the detail. Over what period? Who else was involved? What exactly did you do? What would you do differently now? Real achievements survive that kind of questioning. Borrowed ones tend not to.
Test thinking, not just presence
Cognitive and memory tests are unfashionable in senior hiring, and that is a mistake. Research has consistently found that general mental ability is one of the few reliable predictors of success in a role. A good interview tells you how someone comes across. A well-chosen test tells you something about how they think.
Take at least six references, before the second interview
Most businesses take references at the very end, as a formality, once the decision has effectively been made. I would turn that around. Take at least six, between the first and second interviews, while they can still change your mind.
Make the calls yourself, as the hiring manager, by phone, and keep each one to about ten minutes. Done well, this is one of the most effective techniques I know, and it is a core part of how I run a search for my clients.
Don’t add more interviews. Change what you are watching
When doubts remain, the instinct is to add another round. More of the same weak evidence rarely helps. There are two better options.
The first is a work sample: a test as close to the real day-to-day job as you can make it. If the role is to sell, can you create a setting where they actually have to?
The second is a half or full day in the business. Serious candidates are usually happy to do it. Let them shadow peers and leaders. People relax over a day in a way they never do in an hour across a table, and their natural behaviours start to show, which is exactly what you need to judge fit.
Keep the conversation, but demote it
None of this means abolishing interviews. It means seeing them for what they are: one modest source of evidence among several. The hires that last are rarely made on the strength of the best conversation. They are made by people who looked well beyond it.