6 ms·

Since a lot of Googlers read HN, a question for you: How do you all feel about Google's Launch 'n Kill culture? I did use Trips for travel once and liked it, b

by libria 7y ago

Since a lot of Googlers read HN, a question for you: How do you all feel about Google's Launch 'n Kill culture? I did use Trips for travel once and liked it, but hesitated to get invested in it lest it dies one day (today). I approach most new Google products with the same trepidation. On the other hand, I like Google's start-uppy environment and desire to experiment with anything because it could lead to great products like Gmail. My question is about that circular dependency of a Google product only lasting if it gets popular and a product not getting popular because we think it won't last. If launching is just for annual reviews, and you don't care about the result, I get that but am just curious if any Googlers think the Google Product Stigma needs fixing. Answer as throwaway if needed.

justicezyx 7y ago
Google only interested in billion customer products, sorry, that just what the culture is...
- jfoster 7y ago
  You're right. The exact number might not be right, but the gist certainly is. For that reason, I don't think they should even be exploring products like this. Especially with their volumes of search data, they ought to know whether a product like this will be successful enough for them or not.
evmar 7y ago
[google engineer, not a lot of insight into product decisions] Option one: every product Google makes it supports forever regardless of whether it succeeds or not, where "success" for Google is defined as "billions of users". (Ref all the people laughing at why G+ was still going despite not being popular. [Note: I've read the 'unpopular' thing was disputed; I don't know the actual numbers.]) In this world Google can never attempt a product unless they've verified ahead of time it will become a category-defining product (Chrome, Android, Gmail). Option two: occasionally some products will test the waters and fail. I think it's easy to say at the time of a product failure (or even at launch: see the people who scoffed at iPod and Dropbox) to say that it was destined to fail -- it's pretty much guaranteed to happen on HN because people love to shit on and watch other people shit on others' work, for some sad human nature reason -- but difficult to see which of those guaranteed-to-fail products will actually eventually succeed. So option one mostly means you never launch new products. Big companies faced with this dilemma (the Ciscos or Microsofts or even Facebooks) deal with it by waiting for startups to succeed then buying the result (Whatsapp, Instagram, Github). Famously at Microsoft if you had a good idea the correct way to implement it was to leave, implement as new company, get reacquired. That may become the case at Google too. Instead Google tries to do option two. I think you could argue about the rate of launch/fail in category two but it doesn't bother me too much. I recall a comment from the Netflix founder about how they had too many successful shows and how that is a problem, that they were playing it too safe. I also use emacs as my editor because I know it's old enough that it's not gonna change out from under me. When I care about reliability in this way I choose accordingly. Nothing in life in permanent; I evaluate new Google products like I evaluate random startup products, which is to say they may randomly disappear in a year and I factor that in. For example Keep seems like a nice product and I sometimes use it for grocery lists but I'm not gonna put anything important in it (disclaimer: I have absolutely no insider facts about Keep, that is just my feeling about it based on the previous reasoning).
- pjc50 7y ago
  I think we have to question what "success" means here - all the complaints come from people who were satisfied customers of something with a user base that many startups would be happy to have. Is there no room for "not a world takeover but still fundamentally OK" products?
  - evmar 7y ago
    I'm not sure Google has yet figured out how or why it wants to make such products. I don't understand how apps like Trips or Reader get launched given that they seemed destined for at most mild success, but one possibility is that the people behind them at the time promised some sort of larger success to motivate their funding. True story: I worked on a small Google product where at one point our sales team got a customer interested in a $1m contract, which was a decent amount of money for the scale of our product. But the customer wanted some adjustment in some of the wording. We couldn't get a Google lawyer to even look at the result -- it wasn't worth their time for such a small amount of money -- and I believe lost the customer as a result. To your specific question, I think there is lots of room in the world for fundamentally OK products -- but outside of Google. I love Pinboard for example (well, I don't use it, but I'm rooting for him).
- gravity_123 7y ago
  Re:keep Keep is now a core gsuite app, so while it could also get canned or have a total makeover, it will probably stay, atleast for enterprises, like Google+ continues to.
- libria 7y ago
  I've always preferred Option 2; this is not a criticism of all that entails. In fact, it's no different from all the other startups and side-projects that rise and fall. Google's unique problem is it's all under 1 roof and each product has the Google name on it. So the shuttering of products unfairly get more moaning and press. There's a branding problem here where long term products are in the same group as experiments and Google needs to categorize them to correct public perception. Past approaches were putting things under Google Labs or the Eternal Beta. But it needs to exist separately from the long term survivors like Sheets, GDrive, etc.
Taylor_OD 7y ago
I've heard that the reason launch and kill is so prevalent at Google is that creating a new product is a great way to be promoted while supporting an existing one, even building extensive new and well-recived features, is not. I'm grouping here but engineers that work at Google are likely to be the type of person who is looking to continue to move up the career ladder. Seems like a good option would be to outsource some of these products but that doesn't seem like Googles M.O. either.
- Analemma_ 7y ago
  I just interviewed at Google on Friday, and actually brought this up with my lunchtime interviewer. According to him, Google recognized this problem about a year or two ago, and rejiggered their promotion system to place less emphasis on product launches and more on long-term usage and stability. If that's true (and I only had his word on it), it may just be that they're currently "flushing the pipeline" of all the promotion-fueled products that got started before this change went in, or maybe the change hasn't permeated all the way through the company yet. So it's possible Google's frantic pace of launching doomed products will taper off in time.
- telltruth 7y ago
  I also noticed they seem to shutdown products when the main guy behind it leaves. Typically no one has interest in taking over because it was niche stuff with little chance to go big. Also, original guy would have already milked out credits and there is no career path at Google to be the maintainer in charge. So the sequence of events are: - founding guy leaves - new guy becoming in charge sees career going down drain - makes case for shutdown/half assed merge with other products to get out as soon as possible
wbl 7y ago
Mutt was just fine.

by libria 7y ago

justicezyx 7y ago

Google only interested in billion customer products, sorry, that just what the culture is...

jfoster 7y ago
You're right. The exact number might not be right, but the gist certainly is. For that reason, I don't think they should even be exploring products like this. Especially with their volumes of search data, they ought to know whether a product like this will be successful enough for them or not.

evmar 7y ago

[google engineer, not a lot of insight into product decisions] Option one: every product Google makes it supports forever regardless of whether it succeeds or not, where "success" for Google is defined as "billions of users". (Ref all the people laughing at why G+ was still going despite not being popular. [Note: I've read the 'unpopular' thing was disputed; I don't know the actual numbers.]) In this world Google can never attempt a product unless they've verified ahead of time it will become a category-defining product (Chrome, Android, Gmail). Option two: occasionally some products will test the waters and fail. I think it's easy to say at the time of a product failure (or even at launch: see the people who scoffed at iPod and Dropbox) to say that it was destined to fail -- it's pretty much guaranteed to happen on HN because people love to shit on and watch other people shit on others' work, for some sad human nature reason -- but difficult to see which of those guaranteed-to-fail products will actually eventually succeed. So option one mostly means you never launch new products. Big companies faced with this dilemma (the Ciscos or Microsofts or even Facebooks) deal with it by waiting for startups to succeed then buying the result (Whatsapp, Instagram, Github). Famously at Microsoft if you had a good idea the correct way to implement it was to leave, implement as new company, get reacquired. That may become the case at Google too. Instead Google tries to do option two. I think you could argue about the rate of launch/fail in category two but it doesn't bother me too much. I recall a comment from the Netflix founder about how they had too many successful shows and how that is a problem, that they were playing it too safe. I also use emacs as my editor because I know it's old enough that it's not gonna change out from under me. When I care about reliability in this way I choose accordingly. Nothing in life in permanent; I evaluate new Google products like I evaluate random startup products, which is to say they may randomly disappear in a year and I factor that in. For example Keep seems like a nice product and I sometimes use it for grocery lists but I'm not gonna put anything important in it (disclaimer: I have absolutely no insider facts about Keep, that is just my feeling about it based on the previous reasoning).

pjc50 7y ago
I think we have to question what "success" means here - all the complaints come from people who were satisfied customers of something with a user base that many startups would be happy to have. Is there no room for "not a world takeover but still fundamentally OK" products?
- evmar 7y ago
  I'm not sure Google has yet figured out how or why it wants to make such products. I don't understand how apps like Trips or Reader get launched given that they seemed destined for at most mild success, but one possibility is that the people behind them at the time promised some sort of larger success to motivate their funding. True story: I worked on a small Google product where at one point our sales team got a customer interested in a $1m contract, which was a decent amount of money for the scale of our product. But the customer wanted some adjustment in some of the wording. We couldn't get a Google lawyer to even look at the result -- it wasn't worth their time for such a small amount of money -- and I believe lost the customer as a result. To your specific question, I think there is lots of room in the world for fundamentally OK products -- but outside of Google. I love Pinboard for example (well, I don't use it, but I'm rooting for him).
gravity_123 7y ago
Re:keep Keep is now a core gsuite app, so while it could also get canned or have a total makeover, it will probably stay, atleast for enterprises, like Google+ continues to.
libria 7y ago
I've always preferred Option 2; this is not a criticism of all that entails. In fact, it's no different from all the other startups and side-projects that rise and fall. Google's unique problem is it's all under 1 roof and each product has the Google name on it. So the shuttering of products unfairly get more moaning and press. There's a branding problem here where long term products are in the same group as experiments and Google needs to categorize them to correct public perception. Past approaches were putting things under Google Labs or the Eternal Beta. But it needs to exist separately from the long term survivors like Sheets, GDrive, etc.

Taylor_OD 7y ago

I've heard that the reason launch and kill is so prevalent at Google is that creating a new product is a great way to be promoted while supporting an existing one, even building extensive new and well-recived features, is not. I'm grouping here but engineers that work at Google are likely to be the type of person who is looking to continue to move up the career ladder. Seems like a good option would be to outsource some of these products but that doesn't seem like Googles M.O. either.

Analemma_ 7y ago
I just interviewed at Google on Friday, and actually brought this up with my lunchtime interviewer. According to him, Google recognized this problem about a year or two ago, and rejiggered their promotion system to place less emphasis on product launches and more on long-term usage and stability. If that's true (and I only had his word on it), it may just be that they're currently "flushing the pipeline" of all the promotion-fueled products that got started before this change went in, or maybe the change hasn't permeated all the way through the company yet. So it's possible Google's frantic pace of launching doomed products will taper off in time.
telltruth 7y ago
I also noticed they seem to shutdown products when the main guy behind it leaves. Typically no one has interest in taking over because it was niche stuff with little chance to go big. Also, original guy would have already milked out credits and there is no career path at Google to be the maintainer in charge. So the sequence of events are: - founding guy leaves - new guy becoming in charge sees career going down drain - makes case for shutdown/half assed merge with other products to get out as soon as possible

wbl 7y ago

Mutt was just fine.