I have spent some time this year investigating how hiring is actually working, as opposed to how it is described in textbooks. I wrote up some of what I found in the Wall Street Journal in February and this month in the Harvard Business Review. The main conclusion seems to be that much of what we are doing when it comes to hiring just doesn’t make much sense. Not only doesn’t it look like textbook best practices (which are probably out of date), but it doesn’t conform to basic principles of logic.
This matters because hiring is now the central task in human resources. We spend almost $300 billion each year just on the hiring process in the U.S. It is taking even longer now to get through the administrative processes of hiring (mainly because of more interviewing) than in the past it took to fill jobs, and the costs of extending vacancies that result may be twice the costs of hiring. Despite the perception that it is a candidate’s job market now, they are increasingly irritated by the process employers are using. Those leading today’s hiring practices seem to have forgotten things we’ve known for decades and they are embracing tools with no evidence that they work.
Here are just three of the most serious problems that I see. The first and most important is that we don’t check to see whether our practices are giving us good hires or bad ones. Several surveys report the same finding: Only about one-third of employers, even large ones, collect data to see whether their hiring practices are producing good employees, so they could not find out even if they tried. The common reason as to why they aren’t is because it is hard to measure quality hires. What employers do measure—in many cases, obsessively—are costs per hire and time to hire.
Think about that for a minute: It would be like a marketing department telling you how much an advertising campaign costs and how quickly they put it together without telling you whether customers bought anything. It is particularly damning because checking to see if your hiring practices work is also the best—and only—defense against discrimination charges. It also means that it is impossible to get better.
A second and equally puzzling fact is how thoroughly we ignore internal talent as a source for filling vacancies. This is promotion-from-within or, more generally, internal mobility. Only 28 percent of talent-acquisition leaders report that internal candidates are an important source for filling their vacancies. Virtually all the 66 million hires in the U.S. each year come about by back-filling vacancies, and nearly all of those come about because employees leave for opportunities elsewhere, especially when they cannot advance internally. We are in a hiring frenzy, bailing the boat to keep going, because there is a hole in it caused, at least in part, because we don’t fill those jobs from within.
A third factor, and a bit more difficult to quantify, is that the notion of “culture fit” seems to have emerged as the primary criteria in selection decisions. Yet, we rarely define what our culture is and what a fit looks like, and we make that decision typically based on unstructured interviews conducted by untrained hiring managers. This is a recipe for hiring people “just like me.” At the same time, we fret about diversity and inclusion.
This is where I am puzzled: When I talk to people about these problems in hiring, they virtually always say, “Yes, this is crazy.” So why does it continue? Why is it that we are obsessed with the possibility of things like artificial intelligence when we cannot even accomplish the most basic tasks in hiring, ones that would be necessary if we were ever going to make use of new or more sophisticated tools.
One explanation I’ve heard is that it is just hard to do things differently, especially on something like assessing what constitutes a good hire. There are many dimensions of job performance, and it’s hard to measure some of them, etc. This is a textbook example of making perfect the enemy of good: We will continue to be bad because we can’t be perfect. Even a one-question measure—Would you hire this person again?—is so much better than doing nothing at all.
Another explanation is that we do things the way we do because line managers are calling the shots. They want to pick candidates whom they like, they want to bring in new blood from outside and so forth. If that were true, it would be extremely depressing—that even on this most basic task, where there is a body of evidence going back generations, human resources has no influence.
As you can tell, I am puzzled, and I’d like to hear what you think.