4 ms·

The allocation policy is something that people are not fans of, but it isn't as bad as you make it sound. When I joined Google, I also initially joined a team t

by sxp 14y ago

The allocation policy is something that people are not fans of, but it isn't as bad as you make it sound. When I joined Google, I also initially joined a team that was a bad fit for me. However, I transferred to a completely different team (from enterprise server software to consumer mobile software) after 4 months of near zero productivity, and the new team was a perfect fit. Much of Google's culture is based on being willing to tolerate mistakes and fix them rather getting everything perfect the first time. This applies to both the computer systems (the system should keep running even if part of the network fails) and people. In the latter case, it means the person should be aware that things aren't always perfect and proactively try to adapt and improve the situation. Unfortunately, this chaotic method doesn't always work for everyone and people who want perfect order end up getting stuck in a position they don't like, burn out, and leave. Ultimately, this is bad because it results in the loss of good employees who don't like this chaos, and I'm sorry that it happened to you, varelse. However, I prefer this over a situation where I would interview for a given position and my offer would be based on whether or not I was a good fit for the position that particular position. That method would result in losing people who would be able to fit in well somewhere at Google but not with that specific team.

Jabbles 14y ago
Could you say more about how you arranged to transfer? Others have suggested it is impossible or very difficult, but perhaps they are outliers or victims of older policies. Was it 4 months of trying to be moved, or did you move quickly after trying your assigned role for a while? (I think I would move the other way, into servers :))
- sxp 14y ago
  It was <3 weeks from the time I decided I would be more productive in a different group until the time I was at my new desk. The stages were roughly 1) meeting with my manager and manager's manager to notify them that I wanted to transfer. 2) Meet with other teams to find one that was a good fit. 3) Get all the sign offs from people who needed to sign off. Step 2 took most of the time because I didn't want to make the same mistake that I made the first time. Others who have run into problems and left might be outliers, but it's difficult to come to a solid conclusion based on anecdotes. People who have bad experiences will have stronger feelings and bring up their experiences whenever possible (E.g. michaelochurch who complains about Google every chance he gets) while those who have good experiences like myself don't mention it as much. I only mentioned my case because it was a counterexample to varelse's case, and this is probably true of many others who have had good experiences and don't bother talking about it. Just remember that the plural of anecdote isn't data, and negative anecdotes tend to suffer from selection bias because they have a greater emotional impact upon the person.
  - jjwiseman 14y ago
    I think this is something that varies by office and by manager. I was told that maybe they could let me transfer after a year, but that even a low "meets expectations" score would mean zero chance of being able to transfer.
    - michaelochurch 14y ago
      This is essentially correct. 3.4 is fine, but 3.0-3.1 means you are dead and may not even know it. It's utterly immoral, because a lot of employees at 3.0-3.1 think they're doing fine. Their managers are just giving them low scores to keep them captive. What's amazing to me is that employees don't know their rights. If your manager deliberately interferes with your work performance, that's illegal (harassment law). If a company allows transfer, then exploration of internal opportunities is part of the job. A manager who makes transfer impossible through negative reviews is interfering with work performance and therefore breaking the law. (That, by the way, is why "calibration" happens in secret.) I'd love to see this turn into a huge class action lawsuit against every company (including Google) that does that shit. Serves the fuckers right.
      - chii 14y ago
        > Their managers are just giving them low scores to keep them captive. From reading these comments in this thread, it seems that this score is used to evaluate whether to allow the said employee to transfer. Now, a manager has incentive to keep the best people in his/her team. Is this 3.0-3.1 score something they do so that a good programmer but is somehow "bored" with the crappy work (and face it, there is always crappy work) don't try to all leave? IF that is the case, then there is something wrong.
        michaelochurch 14y ago
        From reading these comments in this thread, it seems that this score is used to evaluate whether to allow the said employee to transfer. HR won't stop transfer with a 3.0-3.1, especially if those are more than a year in the past, but no one will want you. It's the standard ill of closed-allocation shops. When projects compete for people, the result is better projects. When people compete for projects, the result is worse people. Captivity is the most common reason for bad perf. Second place is plain old punishment for things that are political in nature. In third place is "storying", which is when a manager gives bad reviews (and often projects with low visibility and no hope of success) to a good employee to bring him within an inch of his life, puts him through a PIP, and then starts giving him more reasonable scores. The manager gets a story about "rescuing" an "underperformer" and looks like a good boss, but the employee is lucky to transition into the company's lower-middle class, because no one wants a guy who was put on a PIP 5 years ago (having stayed with the company, while under the PIP, reflects on him worse than the PIP itself). The fourth-place cause of bad performance reviews is when the manager takes out someone he fears might be better than him. Somewhere about #13 on that list is actual low performance, because 95+% of real underperformers are so good at underperforming as to be political wizards (they have a lifetime's worth of experience playing politics, since they can't rely on ability) and never end up on the Perf list.

by sxp 14y ago

Jabbles 14y ago

Could you say more about how you arranged to transfer? Others have suggested it is impossible or very difficult, but perhaps they are outliers or victims of older policies. Was it 4 months of trying to be moved, or did you move quickly after trying your assigned role for a while? (I think I would move the other way, into servers :))

sxp 14y ago
It was <3 weeks from the time I decided I would be more productive in a different group until the time I was at my new desk. The stages were roughly 1) meeting with my manager and manager's manager to notify them that I wanted to transfer. 2) Meet with other teams to find one that was a good fit. 3) Get all the sign offs from people who needed to sign off. Step 2 took most of the time because I didn't want to make the same mistake that I made the first time. Others who have run into problems and left might be outliers, but it's difficult to come to a solid conclusion based on anecdotes. People who have bad experiences will have stronger feelings and bring up their experiences whenever possible (E.g. michaelochurch who complains about Google every chance he gets) while those who have good experiences like myself don't mention it as much. I only mentioned my case because it was a counterexample to varelse's case, and this is probably true of many others who have had good experiences and don't bother talking about it. Just remember that the plural of anecdote isn't data, and negative anecdotes tend to suffer from selection bias because they have a greater emotional impact upon the person.
- jjwiseman 14y ago
  I think this is something that varies by office and by manager. I was told that maybe they could let me transfer after a year, but that even a low "meets expectations" score would mean zero chance of being able to transfer.
  - michaelochurch 14y ago
    This is essentially correct. 3.4 is fine, but 3.0-3.1 means you are dead and may not even know it. It's utterly immoral, because a lot of employees at 3.0-3.1 think they're doing fine. Their managers are just giving them low scores to keep them captive. What's amazing to me is that employees don't know their rights. If your manager deliberately interferes with your work performance, that's illegal (harassment law). If a company allows transfer, then exploration of internal opportunities is part of the job. A manager who makes transfer impossible through negative reviews is interfering with work performance and therefore breaking the law. (That, by the way, is why "calibration" happens in secret.) I'd love to see this turn into a huge class action lawsuit against every company (including Google) that does that shit. Serves the fuckers right.
    - chii 14y ago
      > Their managers are just giving them low scores to keep them captive. From reading these comments in this thread, it seems that this score is used to evaluate whether to allow the said employee to transfer. Now, a manager has incentive to keep the best people in his/her team. Is this 3.0-3.1 score something they do so that a good programmer but is somehow "bored" with the crappy work (and face it, there is always crappy work) don't try to all leave? IF that is the case, then there is something wrong.
      - michaelochurch 14y ago
        From reading these comments in this thread, it seems that this score is used to evaluate whether to allow the said employee to transfer. HR won't stop transfer with a 3.0-3.1, especially if those are more than a year in the past, but no one will want you. It's the standard ill of closed-allocation shops. When projects compete for people, the result is better projects. When people compete for projects, the result is worse people. Captivity is the most common reason for bad perf. Second place is plain old punishment for things that are political in nature. In third place is "storying", which is when a manager gives bad reviews (and often projects with low visibility and no hope of success) to a good employee to bring him within an inch of his life, puts him through a PIP, and then starts giving him more reasonable scores. The manager gets a story about "rescuing" an "underperformer" and looks like a good boss, but the employee is lucky to transition into the company's lower-middle class, because no one wants a guy who was put on a PIP 5 years ago (having stayed with the company, while under the PIP, reflects on him worse than the PIP itself). The fourth-place cause of bad performance reviews is when the manager takes out someone he fears might be better than him. Somewhere about #13 on that list is actual low performance, because 95+% of real underperformers are so good at underperforming as to be political wizards (they have a lifetime's worth of experience playing politics, since they can't rely on ability) and never end up on the Perf list.