3 ms·

> Is there some sort of institutional ADD? I work at Google, opinions are my own. Obviously this is my personal take, but I don't think that's what it is. To

by rifung 8y ago

> Is there some sort of institutional ADD? I work at Google, opinions are my own. Obviously this is my personal take, but I don't think that's what it is. To me the issue is the performance/rewards system. Imagine you're someone high up and in charge of 10 projects, all asking for headcount/resources, while you have a limited amount to give. Presumably you're going to give more headcount/resources to the projects that are growing faster. If the differences in growth are substantial, maybe you'll decide to actually just let some projects die so their teams can be absorbed into the other projects with better prospects. Google's performance system is an attempt at being really objective, and it's hard to put a precise value on goodwill lost by discontinuing a product. Not only that, but the effects of discontinuing a product are felt company-wide as opposed to by an individual. Nobody is going to look up who decided to discontinue a product after all.

user5994461 8y ago
> Presumably you're going to give more headcount/resources to the projects that are growing faster. Absolutely not. Your presumptions are misplaced. Assigning resources to growth is usually the most stupid thing to do. By definition the projects that can grow the most are the projects with no customers or revenues, especially new projects that are nowhere, they can only grow. https://xkcd.com/1102/ https://xkcd.com/1102/ . Investing in these is to the detriment of established projects that actually bring revenues and keep the lights on, and companies that neglect to keep the lights on tend to die. Assign resources to the projects where they have the most impact. And business impact should be measured by return on investment (or loss if not invested), this could be almost anything in practice, growing revenues/customers, maintaining existing customers/revenues, not getting fines and lawsuits, reputation control, assets/inventory/cashflow optimizations to save money, etc...
- rifung 8y ago
  > Assigning resources to growth is usually the most stupid thing to do. By definition the projects that can grow the most are the projects with no customers or revenues, especially new projects that are nowhere, they can only grow. https://xkcd.com/1102/ https://xkcd.com/1102/ . Investing in these is to the detriment of established projects that actually bring revenues and keep the lights on, and companies that neglect to keep the lights on tend to die. Well you can consider growth in terms of total users and not the percentage. > Assign resources to the projects where they have the most impact. And business impact should be measured by return on investment (or loss if not invested), this could be almost anything in practice, growing revenues/customers, maintaining existing customers/revenues, not getting fines and lawsuits, reputation control, assets/inventory/cashflow optimizations to save money, etc... We don't disagree. You are talking about what makes sense for a business while I am referring to what makes sense for an individual in an environment where growth is what determines your performance review. That is a bit simplified obviously since of course revenue does matter. My point is that it's exactly this misalignment of what's good for the company vs what's good for individuals working there that is the problem.
ocdtrekkie 8y ago
I guess my question is: Is management there not realizing this is a huge crisis point for them? Clearly, you can make a mistake in performance rewards, but the question is, why hasn't it been fixed? Surely people at the C-level have to have noticed Google is getting a reputation literally defined by sudden product termination, and that the whole company is harmed every time any given product is shut down. I would think this would be a top TGIF question almost every week at this point (because it seems that's how often Google product shutdowns are in the news).
- rifung 8y ago
  > I would think this would be a top TGIF question almost every week at this point (because it seems that's how often Google product shutdowns are in the news). Actually this week (!) was the first time my question was upvoted enough to be asked and it was exactly about this. My impression is that they are aware but willing to trade our reputation for the ability to experiment or be more aggressive in funding more successful products

by rifung 8y ago

user5994461 8y ago

> Presumably you're going to give more headcount/resources to the projects that are growing faster. Absolutely not. Your presumptions are misplaced. Assigning resources to growth is usually the most stupid thing to do. By definition the projects that can grow the most are the projects with no customers or revenues, especially new projects that are nowhere, they can only grow. https://xkcd.com/1102/ https://xkcd.com/1102/ . Investing in these is to the detriment of established projects that actually bring revenues and keep the lights on, and companies that neglect to keep the lights on tend to die. Assign resources to the projects where they have the most impact. And business impact should be measured by return on investment (or loss if not invested), this could be almost anything in practice, growing revenues/customers, maintaining existing customers/revenues, not getting fines and lawsuits, reputation control, assets/inventory/cashflow optimizations to save money, etc...

rifung 8y ago
> Assigning resources to growth is usually the most stupid thing to do. By definition the projects that can grow the most are the projects with no customers or revenues, especially new projects that are nowhere, they can only grow. https://xkcd.com/1102/ https://xkcd.com/1102/ . Investing in these is to the detriment of established projects that actually bring revenues and keep the lights on, and companies that neglect to keep the lights on tend to die. Well you can consider growth in terms of total users and not the percentage. > Assign resources to the projects where they have the most impact. And business impact should be measured by return on investment (or loss if not invested), this could be almost anything in practice, growing revenues/customers, maintaining existing customers/revenues, not getting fines and lawsuits, reputation control, assets/inventory/cashflow optimizations to save money, etc... We don't disagree. You are talking about what makes sense for a business while I am referring to what makes sense for an individual in an environment where growth is what determines your performance review. That is a bit simplified obviously since of course revenue does matter. My point is that it's exactly this misalignment of what's good for the company vs what's good for individuals working there that is the problem.

ocdtrekkie 8y ago

I guess my question is: Is management there not realizing this is a huge crisis point for them? Clearly, you can make a mistake in performance rewards, but the question is, why hasn't it been fixed? Surely people at the C-level have to have noticed Google is getting a reputation literally defined by sudden product termination, and that the whole company is harmed every time any given product is shut down. I would think this would be a top TGIF question almost every week at this point (because it seems that's how often Google product shutdowns are in the news).

rifung 8y ago
> I would think this would be a top TGIF question almost every week at this point (because it seems that's how often Google product shutdowns are in the news). Actually this week (!) was the first time my question was upvoted enough to be asked and it was exactly about this. My impression is that they are aware but willing to trade our reputation for the ability to experiment or be more aggressive in funding more successful products