One Good Engineer Is Better Than Three Normal Ones | Hacker News Reader

One Good Engineer Is Better Than Three Normal Ones

2 pointsjsiaajdsdaa6 comments

Hello,

Why do companies prefer to pay $150,000 each to three normal engineers, when they could pay $450,000 to one good one?

I suspect it is because these companies want to boast about their total number of employees. Perhaps also they are stuck in the mindset of the 1980s factory worker management model where all humans are only capable of operating at the same average speed and error rate.

My assertion is that good engineers can be identified through a proper interview process and enticed with an adequate salary.

If you only hire engineers with average motivation, average speed, and average error rates, the fate of your company will be worse than if you never bothered trying-- money spent, results not delivered.

There is such thing as the bus factor*: he/she can get sick, force-major'ed etc. and that will stall any progress whatsoever, as engineers are people too. Plus can't assign one person to a three different tasks/locations that require attention simultaneously.

*https://en.wikipedia.org/wiki/Bus_factor

I could see having three engineers working on three totally separate projects, but three on the same project usually makes things slower.

i could see your argument from 10 to 100, but 1 to 3 is a practical productivity multiplier. There's also risk of the 1 engineer going to another company that thinks the same way but for $650,000 and now you're left with 0 engineers and no development.

$650,000 is quite the sum, I'd like to know who is paying that :) At least if they leave, you have saved the future salary you would have paid to them, and if they were truly worth their salt, even 3 months of their work is better than 9 months of luke-warm developers (just think of all the CRUD repositories and FooBarFactories you'll have!)

You’re overestimating the signal companies get from any interview processes.

Are you saying that 5 interviews isn't enough of a signal?

Reply on news.ycombinator.com