A large company can afford bad hires. You can shuffle them around or find something they can do.

A small company can not afford that luxury so they have to be more careful in their hiring. A bad hire completely distracts a small team.

Why is simply letting the bad employee go not mentioned as an option?

We had a new hire that didn't work out at a place I used to work at. Fired after 3 months. No hard feelings, he took it well and found a good job a few weeks later. Everyone is happy.

> Fired after 3 months.

Which means you probably went 6 months or more without a role properly filled.

On a small team that's a huge dent. For half a year.

Not to mention the time spent in the hiring process itself, which you now get to repeat.

You can’t have it both ways though. If you have a more selective process than the one that hires-then-fires-after-3-months, you’re _already_ going more than 6 months without filling the role, because by being risk averse to hiring “bad” candidates, you’re passing on good candidates.

So either you pay the cost by trying more people more quickly, or you pay the cost desperately waiting and waiting because you’re so confident your interviewing filter is tuned correctly and that you know better regarding who to hire.

(In a small company, the process of hiring is often quite poor because it’s idiosyncratically controlled by the early employees and founders, and usually reinforces their cultural views and their biases of interviewing, so having a restrictive filter is even more harmful to small companies.)

That’s one of the things. The other is some people are more litigious than others. When that happens, even though they don’t have a case, it can result in distractions —gags, evidence gathering, being called as witness, etc. On occasion companies will rather settle and get the distraction over with than go to court because it drains a small team who are now weary when hiring.

I mean, if you keep a guy who can't do the job, you still end up without a role properly filled.

Google for example does not hire when they think it is not beyond reasonable doubt that the hire is a good one. They rather let a good candidate pass.

This should prove that numbers speak towards the idea that bad hires are very costly. Google's policy is also not to give feedback. 50 percent of their hires had to apply more than once.

I would say large companies do not need that exceptional candidate since they have so many great ones already. Exceptional won't make a difference unless it is executive role or something like that.

> 50 percent of their hires had to apply more than once.

That's one of the reasons why I pass when a Google recruiter comes why way.

(The other is that I don't want to move. I'm no longer a bachelor who can move on the drop of a hat, moving is way to disruptive to my family.)

> “ This should prove that numbers speak towards the idea that bad hires are very costly.”

Why? I do not agree that Google’s hiring practices have anything to do with “bad hire” costs.

Most managers don't want an employee who will be best for the company, they want an employee who will be best for the manager. Easy to get along with, not too likely to cause extra work, etc.

Why do you assume the quoted phrase implies some ridiculous fizzbuzz HackerRank memorization-based hiring process? I agree with Spolsky that generally it's safer to err on the side of false negatives when hiring. I guess the exception would be if truly skilled people in your industry are extremely rare, so you essentially have no choice but to deal with false positives if you want to be able to hire at all.

But none of that has anything to do with the screening process itself, and it's certainly not advocating screening based on bullshit.

The screening process is what generates the true/false positives & true/false negatives.

Any policy about the relative importance of true/false positives vs negatives is a policy about screening processes.

"The last time I came across a difficult problem and overcame it was ..."

"The thing I enjoyed working on the most was ..."

"In five years time, I see myself ..."

"If I could choose to be any animal, I would be a ..."

How large is your company?

The policies in question have worked well for me doing team building in 4 companies, 3 were medium/large ecommerce companies (~1000-8000 employees), one was a 40 person start-up.