or, Why You Should Hire a Potential Disaster
This article presents an angle that might be useful to consider when making hiring decisions, particularly for those interested in building high performing teams.
Some Context You Probably Already Know
While the second half of this post is the more interesting part, let’s establish some context. Here’s a simple model of categorizing your team members:
- Category A employees are the great team members that you love having at your company, you would hire again in an instant, and who clearly pass the “bar”.
- Category B employees are decent, but not great. If you’re honest, you probably wouldn’t hire them again if you could do it over.
- Category C employees are clearly a bad fit for your organization. You’ve hopefully already let them go or are in the process of doing so.
It’s the Category B employees that are the interesting group. In many cases, you end up keeping them at your organization for a long time, for a variety of good and bad reasons (you hope they can move into Category A; you’re unsure whether to classify them into Category A or B in the first place; they’re really nice and hard to fire; you’ve already paid the cost to train them; they produce some useful output).
Category B employees are generally much more costly to your organization than Category C folks because they are likely to be there far longer, while being a slow drag on efficiency and morale.
Try to Avoid Hiring the B’s
The best way to prevent your organization from getting weighed down with B’s is to avoid hiring them in the first place. A few observations may help improve your odds.
Firstly, we know that even the best interview process does not yield a full picture of a candidate, and decisions must be made with incomplete information. We effectively have to ask ourselves whether it’s likely that a candidate will be a worthwhile addition to our company.
There are many ways to ask this question. Some focus on the positive (“Is it likely this candidate will be an “A” employee?”, “Do we estimate the odds are at least 75% that this candidate will be an ‘A’?”, “Do we think there’s a decent chance the candidate is really amazing, an A+?”, etc.).
And some focus on the negative (“What is the risk this candidate will be a disaster, i.e. a ‘C’?”, “What is the chance they turn out to be a ‘B’?”, etc.).
Consider the Range of Candidate Outcomes
Let’s explore some probabilistic distributions we may assign to four candidates, depicting our best guesses about the kinds of employees they will turn out to be.
Candidates 1 and 2 are the easy decisions. We attribute an overwhelming likelihood that Candidate 1 will be an A employee, so obviously we try to hire her. We estimate Candidate 2 is overwhelmingly likely to be a B employee, and thus do not invite her to join.
The next two candidates are more interesting. We believe it is more likely than not that Candidate 3 will be an A employee. However, we imagine there’s a good possibility she ends up a B.
We think Candidate 4 will likely either be a superstar or a total catastrophe.
Whom To Hire?
It can be very tempting to extend an offer to Candidate 3. It is easy to feel that she’ll probably be a good fit, and in case she’s not, she won’t be terrible.
It can also be tempting to reject Candidate 4. Yes, she may be amazing, but she could also be a disaster — and that would be really bad, right?
Remember the Cost of the B
If you remember that B’s are actually far more costly to your organization than C’s, then you quickly see that Candidate 3 is a very risky bet to make, and that Candidate 4 is a great bet.
Hiring a good handful of folks that look like Candidate 3 is a great way to bolster the likelihood you end up with a B-level organization.
If you primarily hire folks that look like Candidates 1 and 4, you will have to deal with a few headaches, but you’re more likely to build a high performing team.
How Do I Determine a Candidate’s Probability Distribution?
Some readers may ask how to construct a probability distribution for a given candidate, as evaluating candidates and detecting meaningful signals in an interview process is challenging. Well, that is an excellent question, and a topic worthy of its own post (or maybe its own book!).
Special thanks to Aneesh Kulkarni, Michel Tricot, Sherif Nada, and Jon Whitney for reviewing a draft of this post.