Never believe stories like this, measuring productivity is an impossible problem. Microsoft spend something in excess of 3 billion dollars trying to quantify productivity in the 90s and they couldn't do it. One of my college professors worked on this exact problem for the Army in the 80s, they created a formal domain specific language for quantifying employee productivity it took them 20 years and several hundred mil…
> Employees always figure out how to abuse productivity metrics. Targets aren't the same things as measures. Measures of productivity are easy, but targets are not. For example, for a sufficiently large group of people (e.g. all young developers at JP Morgan), you can measure their productivity with how many lines of code they write on average. As long as you aren't using lines of code for performance evaluation, the…
I’ve solved problems with 6 lines of code that took me a week to think of. Other days I write 600 lines of code.
I would argue that the former 6 lines I wrote is much more productive than the 600 lines.