New realities of mobile page speed measuring and how to adjust to them

New realities of mobile page speed measuring and how to adjust to them

16th October 2018

These days page speed is prioritized like never before. Of course, such things don’t happen overnight, and there was quite a number of changes recently, which gradually led to it.

Things started changing when mobile-first indexing was rolled out and Googlebot began crawling and indexing pages with the smartphone agent in the first place. But, of course, the biggest change we’ve experienced so far speed-wise was Google’s mobile page speed update, according to which page speed has finally become a ranking factor for mobile devices.

My team conducted experiments, aiming to figure out the correlation between page speed and pages’ positions in mobile SERPs, before and immediately after the update.

What is more, not so long ago Google introduced quite a number of speed-related tools like Lighthouse, Mobile Speed Scorecard and Impact Calculator. It all proves that Google is just going nuts about page speed lately and doing its best to improve mobile user experience.

The two above mentioned updates entailed quite a number of changes in page speed measuring. So, in this post we’re going to talk about these new realities and how to act upon them.

New realities

  1. Speed and Optimization have separate tabs in PageSpeed Insights.

I’m sure you have already noticed a slight update Google has recently made in PageSpeed Insights. The thing is the tool has now two separate tabs for “Speed” and “Optimization” metrics.

When it comes to the Optimization score, it’s evaluated on the basis of the same technical criteria (redirects, compression, minification, etc.) that you used to have in the “Suggestion Summary” section in the previous version of the tool. After analyzing these parameters, the tool will rate your page as good, medium, or low in terms of Optimization.

Speaking of the Speed score, PageSpeed Insights is now categorizing a certain page as fast, average, or slow. The categorization is based on the median value of the following metrics:

  • FCP (First Contentful Paint) measures the amount of time it takes for the first visual element to appear.
  • DCL (DOM Content Loaded) measures how long it takes an HTML document to be loaded and parsed.
How PageSpeed Insights changed in 2018
How PageSpeed Insights changed in 2018
  1. Google incorporates data from the CrUX database

In the light of Google switching from lab data to field data, the way it evaluates your website might not always match your local speed tests. The reason why the discrepancy may occur is that Google now extracts data from CrUX (Chrome User Experience Report). The report collects data from millions of Chrome users: the way they interact with your page, the devises they use, how long it takes your content to load for them, etc.

That is why if the majority of your website users have a slow internet connection, Google may consider it slow even if your website is perfectly optimized. And on the contrary, if a website is poorly optimized, Google may see it as fast due to the fact that most of the users have faster Internet connection or better devises, for instance.

The best way to be “on the same page with Google” is through accessing your CrUX data. Luckily, it’s available on Google BigQuery (part of the Google Cloud Platform). Learn how to get started with CrUX so that you can get first-hand insights from your real-world users.

Note that if you receive an “unavailable” speed score, it means that your CrUX hasn’t collected enough data for your site. But don’t you worry, CrUX database keeps growing constantly, so just check back your speed score in a while.

  1. Optimization score has high correlation with rankings

Earlier in this article, I’ve mentioned the experiment our team has conducted not so long ago. Our mission was to track correlation (if there’s any) between page speed and pages’ positions in mobile SERPs after the update.

After analyzing one million pages ranking on positions 1-30 in mobile search results, we can now firmly state that the correlation between Optimization Score and the position in mobile SERPs is very strong – 0.95.

What is more, an average page ranking within 1 – 30 positions has been improved by 0.83 Optimization Score points after the Speed Update.

However, we’ve hardly spotted any correlation between median FCP/DCL metrics and positions in mobile search results.

Basically, the main takeaway of the experiment is that Optimization Score is what now influences mobile rankings the most. Therefore, working on it should be your primary concern if you’re willing to improve your positions in mobile SERPs.

Speaking of FCP and DCL metrics, although they don’t have any impact on mobile rankings at this very moment, Google is constantly raising standards, and there’s a chance they may become ranking factors in the long run.

Tips for improving Optimization Score

You must have noticed that the PageSpeed Insights tool always supplies you with just a couple of optimization suggestions right after you’ve analyzed your page. However, Google has recently announced that there are all in all 9 possible ways to improve Optimization score, these are:

  • Avoid landing page redirects
  • Enable compression
  • Improve the response time of your server
  • Implement a caching policy
  • Minify resources
  • Optimize images
  • Optimize CSS delivery
  • Prioritize visible content
  • Remove render-blocking JavaScript

For more detailed and technical information don’t hesitate to consult this ultimate guide for improving Optimization Score.

Were “slow” sites really affected by the update?

As you may remember, before the Speed Update went live, Google clearly stated that “it will only affect pages that deliver the slowest experience”. So, within the framework of our experiment, we decided to check how much slow sites were affected by the update if they were at all. By saying “slow sites”, I mean pages that were marked as “Slow” by PageSpeed Insights. So, we’ve compared the sites that Google considers “slow” with the ones that it sees as “fast”.

Surprisingly enough, it turned out that in terms of rankings there was no difference after the update at all. That brings us to a conclusion that it’s definitely not speed that affects mobile rankings. Then we’ve repeated the experiment, but this time we’ve compared sites with high and low Optimization score. Just like before, the experiment revealed that sites with higher Optimization score were least affected by the update.

What it means is that so-called “slow sites” with high Optimization score were not negatively affected by the Page Speed Update. It all proves one more time that, when it comes to rankings, Google seems to prioritize Optimization score rather than page speed (FCP/DCL).

About the author

Aleh BarysevichAleh Barysevich is Founder and Chief Marketing Officer at Link-Assistant.Com, the company that makes SEO PowerSuite (website promotion toolkit) and Awario (social media software) for digital marketing professionals. Aleh is a seasoned SEO expert, speaker at SMX and BrightonSEO, author at Search Engine Land, Search Engine Journal, Social Media Examiner, and more. Connect with Aleh on Twitter @ab80.


Written By
This post was written by an author who is not a regular contributor to State of Digital. See all the other regular State of Digital authors here. Opinions expressed in the article are those of the contributor and not necessarily those of State of Digital.
  • This field is for validation purposes and should be left unchanged.