> It's a good idea to try to quantify the cost of losing an employee. Yes! But this is difficult. But beyond the easily computable costs (e.g., $x that the com

> It's a good idea to try to quantify the cost of losing an employee. Yes! But this is difficult. But beyond the easily computable costs (e.g., $x that the company spends on recruiting / total number of people recruited per year) there are significant costs that involve a lot of assumptions to quantify (e.g., effect of drop in team spirits). Those estimates often get discounted by bean counters exactly because they are based on some personal guesses. In general I find the article somewhat disappointing: it refers to the need to quantify the cost of turnover and IMO does little to help people get defensible estimates. On the other hand it goes on (and on) on why it is important to keep employees. No contention, but I thought the goal was to get to this conclusion by independent, numerical means.