Google Confirms Chrome Consumption Data Used to Measure Webpage Speed
by Tom AnthonyAPR 9th, 2018
During a conversation with Google’s John Mueller by SMX Munich in March, he explained an interesting little bit of data about how precisely Google evaluates site speed nowadays. It possesses gotten a little bit of interest from
people when I stated it at SearchLove NORTH PARK the week after, therefore i followed up with John to clarify my understanding.
The short version is that Google is now using performance data aggregated from Chrome users
who've opted in as a datapoint in the analysis of site speed (and as a sign in relation to rankings). This is a positive maneuver (IMHO) as it means we don’t have to treat optimizing site quickness for Google as a separate
job from optimizing for users.
Previously, it is not very clear how Google evaluates site speed, and it was generally thought to be measured by Googlebot during its visits - a belief increased by the existence of speed
created to replicate how genuine visitors experience a niche site, and so as the task of crawling started to be more complex, it seems sensible that Googlebot might not exactly be the very best mechanism because of this (if
it ever was the mechanism).
In this post, I want to recap the pertinent data for this news quickly and make an effort to understand what this might mean for users.
Google Search Console
Firstly, we should clarify our understand of what the "time spent downloading a
page" metric in Google Search Console is telling us. Many of us will recognize graphs like this one:
Until recently, I was unclear about specifically what this graph was showing me personally. But handily, John Mueller involves the rescue once again with a detailed remedy [login required] (hat tip to James Baddiley from Chillisauce.com
for taking this to my interest):
John clarified what this graph is showing:
It's technically not "downloading the page" but rather "receiving data in response to requesting a URL" - it's not based on rendering the site, it includes all requests made.
that it is:
it is the average over-all requests for that day
Because Google could be fetching a very different group of resources every time when it's crawling your website, and because this graph
does not consideration for anything regarding page rendering, it isn't useful as a measure of the true performance of your website.
Because of this, John points out that:
Focusing blindly on that number doesn't seem sensible.
that i quite agree. The graph can be handy for identifying certain classes of backend concerns, but additionally, there are probably better ways that you can do that (e.g. WebPageTest.org, which I’m a huge fan).
so nowadays we recognize that graph and what it represents, let’s consider the next choice: the Google WRS.
Googlebot & the net Rendering Service
Google’s WRS is their headless browser mechanism based on Chrome 41, which is used for things such as "Fetch as Googlebot" browsing Gaming system, and is increasingly what Googlebot is using when it crawls pages.
we know that isn’t how Google evaluates internet pages because of a Twitter conversation between Aymen Loukil and Google’s Gary Illyes. Aymen wrote up a weblog content detailing it at the time, but the significant takeaway
was that Gary verified that WRS isn't in charge of evaluating site speed:
At the time, Gary was unable to clarify that which was being used to evaluate site performance (perhaps since the Chrome User Experience Report hadn’t been announced yet). It appears as though factors have progressed since that
time, however. Google is now able to tell us a bit more, which can take us on to the Chrome User Experience Article.
Chrome User Experience Report
Introduced in October last year, the Chrome Individual Experience Report “is a consumer dataset of key end user experience metrics for top origins on the net,” whereby “performance data included in the report can be from real-world
types of conditions, aggregated via Chrome users who have opted-in to syncing their browsing record and have usage statistic reporting enabled.”
Essentially, certain Chrome users allow their browser to report again load
time metrics to Google. The statement currently has a public dataset for the most notable 1 million+ origins, though I imagine they possess data for most more domains than are included in the public data set.
I was at SMX Munich (amazing conference!), where along with a tiny group of SEOs I got a speak to John Mueller. I asked John about how exactly Google evaluates site quickness, given that Gary had clarified it was not really
the WRS. John was kind more than enough to shed some light on the situation, but at that time, nothing was published anywhere.
However, since that time, John has confirmed this information in a Google Webmaster Central
Hangout [15m30s, in German], where he clarifies they're using this data along with some other data options (he doesn’t say which, though notes that it's in part because the data set will not cover all domains).
John as well described how Google’s PageSpeed Insights application now
includes info from the Chrome Customer Experience Report:
The public dataset of performance data for the most notable million domains is also obtainable in a public BigQuery project, if you are into that type of thing!
We can’t be sure what the rest of the elements Google is using
are, but we have now find out they are certainly using this info. As I mentioned previously, I as well imagine they are employing data on extra sites than are probably provided in the general public dataset, but this is simply
Focus on users
Importantly, this ensures that generally there are changes you can create to your internet site that Googlebot is not capable of detecting, which are still detected by Google and used mainly because a ranking signal. For instance,
we realize that Googlebot does not support HTTP/2 crawling, however now we know that Google should be able to identify the speed improvements you'll acquire from deploying HTTP/2 for your users.
The same holds true if
you were to use service personnel for advanced caching behaviors - Googlebot wouldn’t take note, but users would. There are undoubtedly other such examples.
Essentially, this means that there's no longer grounds to worry
about pagespeed for Googlebot, and you ought to instead just give attention to bettering things for your users. You still need to focus on Googlebot for crawling requirements, which is a separate task.
If you are unsure
where to look for site speed guidance, then you should look at:
- How fast is fast enough? Next-gen performance optimization - the 2018 edition by Bastian Grimm
- Site Velocity for Digital Entrepreneurs by Mat
That’s all for the present time! When you have questions, please leave your contact details and I’ll carry out my best! Thanks!