Google Algorithm Update May 2021 and Website Speed[ 9 min read ]

Google are making changes to the way they rank websites, based on speed, in May 2021, and in true Google fashion, they’ve made it super difficult to understand and so we’ve endeavoured to translate it into easier to digest information. 

Google has always used their algorithms to rank websites based around user experience and this next change is no different. And although I don’t think they’ve necessarily considered small businesses in their decisions (no change there), there are steps you can do yourself to improve your score/ranking yourself.

During our tests whilst researching, we got a website up from 75% to 96%, within half an hour with two small changes so it’s completely do-able but it may take you longer than that to sort out depending on your current website design, speed management and hosting provider.

We’ve all been there, watching a wheel spin around waiting for a website page to load. It’s annoying. And, I’m sure there have been times when you’ve given up and just left the site.
This is exactly what Google is trying to avoid as this leads to a high Bounce Rate which can indicate a bad experience for the user.

Another reason why speed is important is that the faster a website is, the less energy it’s using (…and we all want to help the planet! Read more about being digitally greener here).

And the less time/resources it takes to load a website is especially important when people are browsing on their mobile.

So, we can understand why Google is striving for faster websites, but not everyone has the budget for super fast hosting, CDNs, or pagespeed management services, and so I feel small businesses will again get the short end of the stick with this Core Web Vitals update.

Alongside the new Core Web Vitals there are other Page Experience metrics in play, which have been around for a while, when it comes to ranking websites; like the site being mobile friendly, the site being safe to browse, the site having an SSL certificate (https), and the site not having intrusive interstitials (i.e. pop-ups that take over the whole screen and the like).

The new page experience metrics are called Core Web Vitals (CWS) and these can be broken down into four sections, although the first one (FCP) doesn’t count towards ranking score as much as the other three.

This is the first point at which the user is aware that the page is loading rather than it just being a blank page. If you’re hanging on and not seeing anything, you’re likely to click off, as mentioned above, and so the page starting to load quickly is a good signal for Google.

This is how quickly the main element on the page, always above the fold, takes to fully load – generally a slider or large banner image but it could be anything.  For elements like big sliders this can often take a while to load and slows down the whole website but, ideally it should take less than 2.5 seconds to fully load for the new Core Web Vitals system.

This is the first opportunity for a user to interact, once the page is loaded, and for the website to respond to that interaction – for example, a link to click, a button on the big slider or banner, or some other kind of interaction which makes the website do something in return.
The time until someone can successfully interact should be kept under 100 milliseconds, ideally.

Have you ever had a webpage shift and change on you whilst you’re looking at it? …usually as you’re just about to press a button and you end up clicking something else entirely because the button has moved down. You think the page is loaded up and you go to press a button but the button shifts to a different part of the page instead.

That is called visual instability and Google isn’t a fan.

But if you have to have something that causes this issue on your website, then they should happen before the first second so your user can continue to browse and interact, unhindered.

There are loads of useful testing sites (see below), including Google’s own PageSpeed Insights. No tool is absolutely perfect, and of course, everyone will have their own favourite (ours is GTMetrix and we hate Google PageSpeed Insights).

When you find one you like, stick with it and only test with that from now on – this will give you the most accurate and consistent score.

• GTMetrix
• Pingdom
• Google PageSpeed Insights
• WebPageTest
• KeyCDN Website Speed Test
• DareBoost
• Web Page Analyzer
• YSlow
• and loads more!

When testing, always make sure the test server is in the same Country every time – ideally, the same country that most of your traffic is from so the score is an accurate representation of a user.

Some suggest testing from further away as well to get an average score, as there is a certain amount of latency with distance, which will affect your score. If you choose to use two locations, ensure it’s the same two locations every time.

Before getting your first score, run the test several times (at least 2-3 to make sure any cache you might have is activated or it won’t be a true representation.
If you clear your cache, you’ll need to do this multiple-testing step, again.

Once you have your first score, test again when you have made changes to the website to ensure your score is within Google’s recommended guidelines. If you’re trying to improve your score, you’ll be able to see whether it’s working or not. Some of the test sites (like GTMetrix) let you save your results so you can see the progress.

Once you have your score, it will be split between speed loading times and the four CWVs which will give you a clue as to which bit you might need to work on to improve your score.

For example, if your score is too high for LCP then you will need to make the main element on your page (above the fold) load faster either by reducing the size or by implementing cache or lazy loading.

If your score is too high for CLS then you might want to remove the issue or again, make it load faster.

If your overall speed score is too high then you may need to look at other hosting options or introduce some speed management tools like browser caching or lazy loading.

Yes, you could; if you think you have elements on the page (especially above the fold) that are unnecessary, remove them or move them to a more relevant page.

For example, do you have 6 large slides in your banner slide-show? Could it be changed to a smaller 1 image banner instead?

Do you have fancy animations that look whizzy but don’t add any functionality to an already heavy-content page? We had this and so we removed the animation as, in our opinion, it’s much better to have a website that’s ranking well than one that is animated.

If you don’t understand your results, then you can always book in for a Digital Power Hour with us where we can go through them with you and make suggestions.

Or, of course, we can do the testing and improvements for you if you want us to. Get in touch below for an estimate based on your website.