Mobile site load speed in search engine optimisation

Many marketers are taking the initiative to investigate the state of search engine optimisation of their website and look for ways to improve findability. Quite often, they start by examining the site's load speed. When you are unfamiliar with the subject, you can end up with wrong information and wrong conclusions right from the start.

The technical performance of a website and its loading speed affect SEO results. When measuring the loading speed of a website, it is easy to be misled into following the wrong figures or data sources. And when the measurement result looks poor, people start to fix things that might not need to be fixed.

One of the biggest misinterpretations is when using Google's PageSpeed Insights tool as a data source and looking at site performance on mobile devices.

PageSpeed Insights and its results

In terms of site performance and load speed, PageSeed Insights can display results in two ways, results based on field data and results based on lab data. Field data is based on the experiences of real users of the site... If field data is available, it will be displayed first in the report.

If at the beginning of the results report, "Discover what your real users are experiencing" or "Discover what your real users are experiencing" has the text "No data", the test will not show the field data, only the results of the lab test.

Mobile device testing and laboratory data

At the bottom of the opening screen of the "Performance" or "Efficiency" section of the PageSpeed lab test, you can see what parameters the test was performed with. It says, for example, "Slow 4g throttling" or "Slow 4G simulated throttling" is enabled. This means that the test artificially lowers the network speed to the average speed in Finland more than ten years ago.

In the lab test, PageSpeed uses a mobile assessment to underestimate the conditions of the user environment as they are likely to be for real users of the site. In this case, the tool will not show results that match the experience of real users.

This is why the mobile test gives a result that is probably quite different from the experience of real users, if the majority of the site's users are from Finland. When testing the site on your own phone, you may often find that the page does not take as long to load as the test indicates.

Read more laboratory test vs field data from web.dev.

So what is PageSpeed used for?

The PageSpeed lab result is not used by Google for SEO evaluation. It is intended to provide guidance for site development, especially when creating global sites that need to take into account all types of mobile devices and interfaces. In Europe, and especially in Finland, devices and interfaces are very different from those in emerging countries, for example.

If there is enough real user data, PageSpeed will show values based on field data. These values tell you the true performance of your site and can be used by Google for SEO evaluation.

The role of download speed in search engine optimisation

When Google evaluates the technical performance of a website, the download speed is one factor within it. The impact of load speed on search engine optimisation is not well understood. After researching the issue, we have come to the conclusion that it is an on/off value. Either that is the case, or it is not. In other words, if the loading speed is good enough, there is not necessarily any further benefit from a search engine optimisation perspective to speeding up the site. Of course, the fastest possible site will always have a positive impact on the user experience.

The challenges of different analytical tools

The web is full of different measurement, auditing and testing tools, some of which are intended for professional use and some of which are intended to provide data that can be used to argue for the sale of SEO services, for example. If you don't know how the metric works, it is unlikely to be useful for website development.

Proven and high-quality metrics such as GTmetrix, Pingdom and Webpagetest.org, in addition to Pagespeed Insights, provide reliable data to measure the performance of your website. Be sure to check where in the world the test server is located and whether there is any setting that artificially slows down the speed.

In many cases, search engine optimisation should focus on factors other than the technical speed of the site, unless there are real problems with loading speed. Site speed is important for user experience, conversions and SEO, but you need to assess it properly and consider whether it's worth developing. One way to get more out of it is to focus on developing the content of your site.