Even for Search, Latency Matters
Even for a relatively non-latency sensitive application, such as Google search, there are measurable business implications. Google’s own experiments “demonstrate that slowing down the search results page by 100 to 400 milliseconds has a measurable impact on the number of searches per user of -0.2 percent to -0.6 percent, says Jake Brutlag, Google “Web search infrastructure” staffer.
Basically, that means 0.2 percent to 0.6 percent fewer searches for changes under half a second of delay. For a company that makes its money serving up ads on search results pages, that means a real loss of income-earning potential.
Furthermore, users conduct fewer and fewer searches the longer they are exposed to the delay. Users exposed to a 200 milliseconds of delay since the beginning of the experiment did 0.22 percent fewer searches during the first three weeks, but 0.36 percent fewer searches during the second three weeks.
As you would expect, people avoid experiences that are unpleasant.
Users exposed to a 400 ms delay did 0.44 percent fewer searches during the first three weeks, but 0.76 percent fewer searches during the second three weeks.
Even if the page returns to the faster state, users who saw the longer delay take time to return to their previous usage level. Users exposed to the 400 ms delay for six weeks did 0.21 percent fewer searches on average during the five week period after we stopped injecting the delay.
While these numbers may seem small, a daily impact of 0.5 percent is of real consequence at the scale of Google Web search, says Brutlag
Even for a relatively non-latency sensitive application, such as Google search, there are measurable business implications. Google’s own experiments “demonstrate that slowing down the search results page by 100 to 400 milliseconds has a measurable impact on the number of searches per user of -0.2 percent to -0.6 percent, says Jake Brutlag, Google “Web search infrastructure” staffer.
Basically, that means 0.2 percent to 0.6 percent fewer searches for changes under half a second of delay. For a company that makes its money serving up ads on search results pages, that means a real loss of income-earning potential.
Furthermore, users conduct fewer and fewer searches the longer they are exposed to the delay. Users exposed to a 200 milliseconds of delay since the beginning of the experiment did 0.22 percent fewer searches during the first three weeks, but 0.36 percent fewer searches during the second three weeks.
As you would expect, people avoid experiences that are unpleasant.
Users exposed to a 400 ms delay did 0.44 percent fewer searches during the first three weeks, but 0.76 percent fewer searches during the second three weeks.
Even if the page returns to the faster state, users who saw the longer delay take time to return to their previous usage level. Users exposed to the 400 ms delay for six weeks did 0.21 percent fewer searches on average during the five week period after we stopped injecting the delay.
While these numbers may seem small, a daily impact of 0.5 percent is of real consequence at the scale of Google Web search, says Brutlag
Tags: application latency, Google, search, search latency
This entry was posted
on Friday, March 19th, 2010 at 4:13 pm and is filed under Applications, News.
You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
Heya i’m for the first time here. I came across this board and I in finding It really helpful & it helped me out much. I hope to present one thing back and aid others such as you helped me.
Sorry I forgot that SNA to OGG flight. There is one!
Its like you learn my mind! You appear to know a lot about this, such as you wrote the guide in it or something. I think that you could do with some percent to power the message home a bit, but instead of that, this is great blog. An excellent read. I’ll certainly be back.