How startups can move beyond the old fashioned job interview

Signal and the noise
Francois Badenhorst
Deputy editor
BusinessZone and UK Business Forums
Share this content

Here’s the problem with job interviews as you know them: we all think that we’re a better judge of character and aptitude than we actually are.

We suck at these things through no fault of our own, really. It’s because our brains are damned liars. As we float through life, the brain constantly scans the perimeter for patterns. While that may seem harmless, it does have the tendency to find meaningful patterns in the meaningless. It’s a cognitive bias the American science writer Michael Shermer calls ‘Patternicity’.

In Prehistoric times, this had its uses as a survival tactic. As Shermer explains it: “The cost of believing that the rustle in the grass is a dangerous predator when it is just the wind is relatively low compared with the opposite.” The brain, it seems, likes to hedge its bets.

Patternicity, similar to more physical leftovers like our tailbones, is a vestige from a different time. It retains an acute power, however, because of how it still affects our interactions with others.

Consider the job interview. A familiar phrase when hiring is ‘getting a feel’ for the candidate. Interviews resemble winding conversations that, the theory states, will allow the savvy interviewer to glean the insight they need.

What form the structure takes - a test, or just set questions - isn’t as important as the structure itself.

There’s a problem here, though. Jason Dana, a professor at the Yale School of Management, tallied these issues comprehensively in a recent (really good) op-ed in the New York Times. Titled The utter uselessness of job interviews, Dana argued “people have no trouble turning any information into a coherent narrative”.

“This is true when … the information is incorrect. And this is true, as in our experiments when the information is random,” he wrote. “People can’t help seeing signals, even in noise.” Dana recounts the exhaustive experiments to back these assertions in NYT article.

At BusinessZone, we’ve written extensively about the challenges of hiring. It’s axiomatic that in a business’ infancy, new hires can make or break the company. In a small team, a hire can radically alter the frequency. As Olly Culverhouse, founder of the e-signature startup Signable, told BusinessZone: “Every new hire is a sizeable percentage change in the number of people at a small company.”

The co-founder of Y-Combinator Paul Graham has three golden rules when it comes to hiring for startups. Number one is rather blunt: “Don't do it if you can avoid it.” Of course, it can’t be avoided forever.

At seven people - and hiring a few more positions, too - Culverhouse has found success eschewing the traditional job interview.

The inherent risks often lead founders to turn inward and hire from their network. It’s seen as an easy way to mitigate the risk of a bad hire. As the entrepreneur Kevin Holler explained, though: it’s not that simple and a mistake could come at a personal cost.

It’s something Signable’s Culverhouse has thought about, too. Although his first hire was a good friend (and it worked out fine), he has subsequently avoided just opting for the network based, personal interview format of hiring.

At seven people - and hiring a few more positions, too - Culverhouse has found success eschewing the traditional job interview. After an initial, very informal chat (that doesn’t broach any technical topics), the interview process moves to a test.

Candidates are given a relevant task and a deadline of one week to complete the test. For this, they are paid £250. “It’s a great way to test how good the candidate’s skills are in a fixed, controlled way,” says Culverhouse. “We set them a task and deadline and we can see how competent they are within that.”

For a marketing position, Culverhouse sets a task of building a basic marketing strategy for an industry given a certain budget. For a technical hire, the test centres on using Signable’s development tools to build something very small that uses them.

This kind of structure, according to Dana’s research, is a more reliable indicator of a candidate’s future success. And, perhaps more importantly, it's fairer. What form the structure takes - a test, or just set questions - isn’t as important as the structure itself.

“The hiring process is one that I spend quite a bit of my time on. It’s so vital to get the right people in, I know it’s a cliche,” he says. “It’s definitely something we can keep doing from a scalable point of view because it doesn’t take much longer than a normal recruiting process and there’s a lot less risk. And also the person who is potentially joining us, gets a deeper insight into how we work. It helps us attract better people.”

And, as Culverhouse points out, he wanted Signable to cast the widest net, especially since he didn’t have the best network to begin with. “I didn’t want to just hire friends because we had no one else. So this was quite a good balance, really.”


Please login or register to join the discussion.

There are currently no replies, be the first to post a reply.