How to Hire
Henry Ward

I agree with most of this.

> Hiring means we failed to execute and need help

I don’t think that this is always true, but I agree strongly with the sentiment. A hundred programmers crammed into an open-plan office is not “success”. It’s inefficient. Something I’ve noticed is that most companies are reactive and that they promote people into management for the same reason that they fire people: the person is no longer able to get all the work done that is thrown their way. It seems, then, to depend on political skill which outcome a person gets. If you’re politically favored, you get a team when you’re swamped. If you’re politically out of favor, you get demoted or fired for the same thing (more work than you can handle).

That said, while I like this idea as an antidote to executive egos and headcount fetishism, I also think that it misses something. Success means progress, and that means drawing on new expertise as well as expanding operational capacities that can’t yet be automated (for regulatory reasons, or because it’s prohibitively difficult). Hiring is second-best to not having to hire in order to expand, but often necessary and not really a sign of anyone’s failure.

> We underestimate the most effective employees and overestimate the ineffective ones.

What’s missed in the graph with the dots is the possibility that the perception is the *reverse* of actuality. I’ve seen that many times.

> She is in the wrong role. [cited, with doubt, as a reason why a good employee might underperform]

Actually, that’s quite a common reason for a good employee to underperform. Constitutional low-performers exist, but they’re really rare (less than 1%, probably less than 0.25%) in high-end knowledge work unless you’re awful at hiring. I’m not saying that a 3.5+ GPA in a STEM major from a good college means that a person is necessarily brilliant, but someone with that history is by demonstration not a *constitutional* low performer. Usually, it *is* a problem with the context. It might not be a problem that you can fix, and you might have to hire a good person who landed in the wrong place because too much damage has already been done, but that’s a separate issue. Most often, “wrong role” is the correct diagnosis.

> False Positives are ok, False Negatives are not

For non-managerial positions, I agree. For managerial positions, false positives do a lot of damage and are much harder to spot, because you’re typically talking about people who are really good at politics and manipulating perceptions. Hire a worker who turns out to be a dud, and you can package him out with 3 months’ severance and everyone’s happy. Hire a bad manager, and you’ll lose people, have bad decisions get integrated into the culture, and possibly even wreck the company. Bad managers are usually not just unskilled but unethical, and will often have 2 or 3 good employees take falls for them (likely getting fired) before they themselves go down. By that time, in addition to the 2–3 bad fires, you’ve had several good employees quit. So you might lose 10 people, over 3 years or more, before you realize that you made a mistake and get rid of the bad manager.

> Hire for Strength vs Lack of Weakness

I agree absolutely. At Valve, they talk about the “beaten wife syndrome”, which seems to be an industry-wide affliction considering that inoffensiveness and reputation trump competence, leaving professionals (especially in software) afraid to work in a way that’s congruent with how they actually think and what they actually want. Companies seem to want to hire people that no one can say anything bad about, but that means that you’re hiring the same person, over and over. This is fine if your target is reliable mediocrity, but the best people want to aspire to something better.

> Always pass on ego

Disagree. I also think that it’s possible to have humility and an ego, and that this is probably the best combination. For example, I know what I’m good at, and that there are fields where my knowledge is more extensive and my ideas are better than that of almost anyone else. I also know that there are vast sectors of human knowledge in which I know very little. “Leader of Men, humble before God”. (Here, I’m using “God” rhetorically rather than theologically.) It’s important to have an accurate self-perception and know where you rank, but also to recognize that you’re often *not* as knowledgeable as the person who’s been studying a topic for years, and to listen rather than talk, when that is the case.

The thing about *ego* is that it’s pretty much impossible to become a great software engineer if you don’t have an ego. You have to change jobs, as soon as you catch a whiff that you’re getting stuck with work that you won’t learn anything from. You have to inject yourself into conversations, read papers that might seem “too advanced” for you and that your boss certainly wouldn’t appreciate you reading on the clock, and elbow your way on to the best teams and projects so that you keep learning. Good programmers aren’t that way just because of natural talent. It requires getting yourself on the best projects, and you can’t do that without the “ego” to believe that you belong on them (and to leave jobs, and blow off work, that does nothing for you).

The best combination is someone who is arrogant/egotistical enough to break through the impostor syndrome, to learn without being daunted, to fight for herself when she is right, and to learn “on the clock” and take creative risks without asking for permission… but also humble enough to listen more than she talks, to learn from other people, and to step down when she’s not the most knowledgeable person in the room.