Some other perspectives: - Salary is VERY important for most engineers - most of us are making incomes where making a little more money might be important. Tha

Some other perspectives: - Salary is VERY important for most engineers - most of us are making incomes where making a little more money might be important. That said, consider the incentives that drive the decision makers (managers) who decide on salaries. The managers who say "Yes we can keep these people but no we don't have to pay them more" are much more likely to succeed than the ones who say "I need more money for the same headcount". Unless you've put serious institutional capital in the bank in the form of previous success, you won't win this argument with higher management. (By the way, I agree with the general premise that turnover is really expensive and most companies, including their upper management, don't take it seriously). - I think that this is another example of the "Hard to measure, easy to measure" divide in otherwise scientific management that defines most companies today. It is easy to measure salary. In fact, you are usually required to report it in most accounting schemes. Measuring cost of employee acquisition is easy for some parts only. On-boarding and training costs can be relatively well quantified. But what about lost productivity? What about lost innovation? Trying to measure "what isn't there" has never been easy. Managers will make decisions based on the hard data at hand. There is almost never enough hard data to justify greatly increasing their salary, and most managers won't go out on a limb (lacking hard data) to try and justify this. I can't say I'd do differently in their situation...