Implement a ranking system and standardized tests. Count lines of code and apply it as a bonus multiplier. Make sure to only measure things directly assigned to a person, while ignoring 'off-the-book' things like bringing new hires up to speed, code review, etc. Devote a large portion of the bonus to a 'face-time in meetings' metric. Put everyone on an artificial bell-curve measuring whim in exquisite detail.
20% of staff-time should be taken off anti-pagerank gaming efforts and redirected at anti-bonus-metric gaming.
Because clearly the idea that you have high and low performers is brain dead, and therefore the only way to measure that stuff is stupidly, duh.
Come on. In any organization of that size there are people you really want to hang on to. People about whom the question "How would I replace that person?" is harder to answer than for others. My only point is if the stated goal is to retain people, put the resources toward those you actually want to retain. If Google is so exceptional that they really don't have that issue, then I am totally impressed.
Of course the other possibility is that Google is without constraint when it comes to compensation resources, which would also be completely foreign to my experience and invalidate my point.
I think the problem is simply their size makes it very difficult to identify their top performers. There is no objective measure that can be applied across the entire organisation, so they have to rely on humans to identify the top performers. Their top 10% of staff would come to 2,300 people (assuming the 23,000 mentioned elsewhere is correct), which means no single person in the organisation can identify the top people (and probably no single person knows all of the top people).
So you need to rely on mid-level managers to identify the top performers, but each mid-level manager is likely to be biased towards their own staff and their area of responsibility. And of course the mid-level managers may not be in the top 10% and so may be unable to identify who the star performers are anyway. So then you try to add some kind of normalisation scheme, maybe you take the top 10% from each department. Except that doesn't account for one department being better/more important than another. So you have to add that weighting in,... and you end up with some complex system that no-one really understands or agrees with.
Add to that the evidence that performance related pay isn't always beneficial and this becomes a very difficult problem.
And all of that extra work to try and identify the top 10% of staff would probably wind up costing the company more than a simple across-the-board raise does.
Counting the lines of code an engineer has written is a terrible road to go down. I can write 500 lines of good code, or I can abstract it into a 5,000 line heap of unmaintainable garbage.
Comments
Implement a ranking system and standardized tests. Count lines of code and apply it as a bonus multiplier. Make sure to only measure things directly assigned to a person, while ignoring 'off-the-book' things like bringing new hires up to speed, code review, etc. Devote a large portion of the bonus to a 'face-time in meetings' metric. Put everyone on an artificial bell-curve measuring whim in exquisite detail.
20% of staff-time should be taken off anti-pagerank gaming efforts and redirected at anti-bonus-metric gaming.
Because clearly the idea that you have high and low performers is brain dead, and therefore the only way to measure that stuff is stupidly, duh.
Come on. In any organization of that size there are people you really want to hang on to. People about whom the question "How would I replace that person?" is harder to answer than for others. My only point is if the stated goal is to retain people, put the resources toward those you actually want to retain. If Google is so exceptional that they really don't have that issue, then I am totally impressed.
Of course the other possibility is that Google is without constraint when it comes to compensation resources, which would also be completely foreign to my experience and invalidate my point.
I think the problem is simply their size makes it very difficult to identify their top performers. There is no objective measure that can be applied across the entire organisation, so they have to rely on humans to identify the top performers. Their top 10% of staff would come to 2,300 people (assuming the 23,000 mentioned elsewhere is correct), which means no single person in the organisation can identify the top people (and probably no single person knows all of the top people).
So you need to rely on mid-level managers to identify the top performers, but each mid-level manager is likely to be biased towards their own staff and their area of responsibility. And of course the mid-level managers may not be in the top 10% and so may be unable to identify who the star performers are anyway. So then you try to add some kind of normalisation scheme, maybe you take the top 10% from each department. Except that doesn't account for one department being better/more important than another. So you have to add that weighting in,... and you end up with some complex system that no-one really understands or agrees with.
Add to that the evidence that performance related pay isn't always beneficial and this becomes a very difficult problem.
And all of that extra work to try and identify the top 10% of staff would probably wind up costing the company more than a simple across-the-board raise does.
Funnily, apart from the LoC as a bonus multiplier, my friends at Google complain about exactly those things -- especially code reviews and new hires.
I guess the OP is also a googler so knows the pain point exactly!
Counting the lines of code an engineer has written is a terrible road to go down. I can write 500 lines of good code, or I can abstract it into a 5,000 line heap of unmaintainable garbage.
That was the point ;)
He was being ironic. Are you being doubly ironic? Am I not seeing the iirony?