• by Gord Hotchkiss , Featured Contributor, August 9, 2013

In all my years in business, the one thing I found consistently difficult was hiring good people. We spent a lot of time honing our screening skills, but I sometimes suspect we would have been just as far ahead by flipping a coin.

Over time, we found we achieved pretty good success rates with our lower-level hires, but the one area where consistent success eluded me was in our management recruitment. It seemed that the more senior the position, the worse our track record was. We had a few outright disasters.

When it comes down to it, hiring someone is making a prediction. You examine the evidence and try to foresee if that person will perform at an acceptable level in the position you have vacant. And, as I said in my last column, we humans don’t tend to be very good at making predictions. The more there is at stake in the position to be filled, the worse the consequences if our predictions are faulty. In looking at our past management hires, I realize that it wasn’t that our predictive powers were any less effective, it’s just that the pain of being wrong was more acute.

So, it was with some reassurance and more than a dollop of schadenfreude that I learned that Google has had exactly the same problem. That’s right, Google -- the same company that has a zillion brilliant engineers working on every problem known to mankind. But those engineers have to come from somewhere, right? Someone has to hire them. And there, ay, there’s the rub!

In a recent interview in the New York Times, Laszlo Bock, senior vice president for people operations at Google, confessed that Google has tweaked, and, in some cases, massively overhauled its recruitment process.  Take, for example, Google’s famous early predilection for college G.P.A.s. According to Bock, based on actual performance, “G.P.A.s are worthless as a criteria for hiring, and test scores are worthless — no correlation at all except for brand-new college grads, where there’s a slight correlation. Google famously used to ask everyone for a transcript and G.P.A.s and test scores, but we don’t anymore, unless you’re just a few years out of school. We found that they don’t predict anything.”

Google has also slowly backed away from its ironclad requirement that every hire have a degree. Bock revealed, “The proportion of people without any college education at Google has increased over time as well. So we have teams where you have 14% of the team made up of people who’ve never gone to college.”

Sometimes, interviewers fall into the trap of over-playing their own cleverness and “expertise.” We spend more time trying to stroke our own ego by staging an impromptu show of power during the interview than in really listening to what the interviewee is saying. Google found that tricks like brainteasers, while they may make the interviewer feel clever, are worthless in screening out duds. The much less flashy but tried-and-true list of standardized behavioral questions (“Give me an example of when you…”) is a far better predictive indicator.

Finally, Bock admits that screening for leadership positions is the most difficult challenge, because leadership is something that defies easy definitions. “We’ve found that leadership is a more ambiguous and amorphous set of characteristics than the work we did on the attributes of good management, which are more of a checklist and actionable.” So you can ask questions, probing for effective leadership, but because leading people tends to fall into the category of ill-defined problems, you often have to do the best job you can in the hiring process, and then track performance religiously. In this case, “slow to hire, quick to fire” is a good principle to follow.

I found Bock’s last words, on the role of Big Data in management decisions, including those involving people’s performance, revealing: “Big Data — when applied to leadership — has tremendous potential to uncover the 10 universal things we should all be doing. But there are also things that are specifically true only about your organization, and the people you have and the unique situation you’re in at that point in time. I think this will be a constraint to how big the data can get because it will always require an element of human insight.”