I'm an Xoogler and was promoted when I was there. (see sibling comment for details).
The way to get promoted at Google seems to be to play a game where you tick all the boxes for performance at the next level and have the right people write for you. In some ways, that helps the company (working across teams, for example). In other ways, it may hurt the company (launching potentially redundant products is seen by committees as being more valuable than incrementally improving existing products). I think that a lot depends on who writes recommendations for you in your promo packet (and that's pure politics)
In my case, I was basically gathering requirements and helping other teams integrate with an internal product. So I was perfectly positioned for promo. I was an L5, and talking to a lot of senior folks in other teams (sr. staff, director, vp) who were willing to write for me. I'm pretty sure having a VP who knew me and could write about me really made the difference.
> I think that a lot depends on who writes recommendations for you in your promo packet (and that's pure politics) ... I'm pretty sure having a VP who knew me and could write about me really made the difference.
It sounds like that VP knew you because of the work you were doing, not because you bought him dinner or something. In that case, it's not pure politics, your work was apparently important and impacted a lot of people.
In this thread it sounds like a lot of people think promotion (and work performance in general) should be measured by purely technical contributions only, which is not realistic. That kind of work is important and makes sense for entry level work, but that's not how big projects get done.
That's true, he knew me because of the work I was doing on my 20% project. This was a small stealth-mode thing which was a pet project of his. So I mainly just lucky. Meanwhile, the guy who sat behind me, worked much harder, came in earlier and left later, and turned out mountains of high quality code never got promoted (he retired).
But Google prides itself on being a meritocracy, when really it is just another case of "who you know" is as important as "what you know"
(my guesses from reading this thread) The work was important, but the GP was _the face_ of that work only. Many other people collaborated on shipping that work - the VP only knew GP. The issue people are pointing out is that this incentivizes people gravitating towards such roles which are high in. "visibility".
Don't know how it is now, but most of my promos at Microsoft (in 00s) were unexpected by me. I just kicked ass, and my managers delivered the goods. I liked that system. That's how it's supposed to work. :-)
Some other people have mentioned hiring people defensively so they can't compete. I don't know how true that is, but if it is... then making a redundant project means you need to be retained so you don't write clones for anyone else.
That's not a problem with the process, that's a problem with what the senior engineers value (making themselves feel smarter than their aspiring peers, not producing valuable products)
The way to get promoted at Google seems to be to play a game where you tick all the boxes for performance at the next level and have the right people write for you. In some ways, that helps the company (working across teams, for example). In other ways, it may hurt the company (launching potentially redundant products is seen by committees as being more valuable than incrementally improving existing products). I think that a lot depends on who writes recommendations for you in your promo packet (and that's pure politics)
In my case, I was basically gathering requirements and helping other teams integrate with an internal product. So I was perfectly positioned for promo. I was an L5, and talking to a lot of senior folks in other teams (sr. staff, director, vp) who were willing to write for me. I'm pretty sure having a VP who knew me and could write about me really made the difference.