4 ms·
Genuine question. Is there any verifiable, quantitative data to substantiate what you wrote or is it all a HackerNews urban-myth? I am genuinely curious becaus
by shiven 12y ago
Genuine question. Is there any verifiable, quantitative data to substantiate what you wrote or is it all a HackerNews urban-myth?
I am genuinely curious because I hear that argument thrown around a lot, anytime non-monetary compensation is being discussed among the HN crowd.
- kooshball 12y agoIf you're talking about the 5x revenue there are certainly a lot of public data on that. http://www.wolframalpha.com/input/?i=aapl+goog+msft+facebook http://www.wolframalpha.com/input/?i=aapl+goog+msft+facebook Look for revenue/employee.
- rdl 12y agoAnd that's with LOTS of overhead people, communications complexity, etc. If you strip out just the developers at a big company which is very profitable and product focused, you'll get even more insane revenue-per-employee figures. It's depressing/unpolitic to say, but the "free meals, shuttles, etc." are intended for the developers, management, and other highly-in-demand people. It's probably overall worse for corporate culture to set up multiple tiers of employees, some of whom don't qualify for those perks, but a company with huge numbers of easily-recruited, low-value-add employees (e.g. walmart) tends to have crappy perks; a company with a small number of key employees with absurd productivity and profitability per employee (hedge fund?) tends to have amazing perks. (That startups tend toward binary outcomes based on the productivity of the first 5-10 people is probably not in dispute here, I assume and hope.)