Or at least it’s not as important as you think it is.
Last week I was asked by a prospective client to use the score from Google’s Page Speed Insights tool as the primary KPI for a possible SEO campaign. According to this client, their website was struggling to appear for their target key terms because the site was too slow.
They said this with absolute certainty: “Well obviously the main thing holding the site back is the site speed…”.
The site itself has about 3 hub pages for their main services, and a blog full of rehashed Linkedin articles. The service pages were in no way optimised for their target keywords. There were no headers, no meta data, a little content. There was clearly no defined strategy behind the blog post, just a mish-mash of random thoughts that didn’t support the site’s pillar pages in any way.
There was a clear need to appear for local searches, but there were no geo-specific pages or mention of any location beyond the address in the footer.
But yeah, of course, the site wasn’t doing well because of its slightly below average Page Speed Insights score.
Misconceptions about Page Speed
This is a real bugbear of mine, one I find myself getting increasingly frustrated about. So many website owners, agencies, SEO, developers, marketing managers, every bloke and his dog are OBSESSED with Page Speed/Site Speed/Load Speed/Whatever Speed you want to call it.
Before I get into this, yes of course it’s important that your site isn’t slow. It’s just obvious. Users will leave your site if it doesn’t load quickly etc etc etc. If your site loads like it’s on a late 90s dialup modem, of course you need to sort yourself out. This is just common sense.
But for the vast majority of websites in existence, improving your Page Speed score will have a marginal impact on your site’s SEO performance, if it has any impact at all.
Now, even I am nowhere near clever enough to fully understand the nuts and bolts behind all of this. A lot of great technical SEO people can provide a better explanation.
But here’s a few points as to why there are few very big problems with using Page Speed as a KPI.
- When you run a Page Speed Insights/Lighthouse/GTMetrix/Pingdom test, you are measuring how quickly your specific machine can process that web page on your specific internet connection from your specific location.
- Many external factors influence your Page Speed score beyond the website itself. The internet is a big mess of flying internet stuff, and all of this factors into how quickly you might load a website at any given time.
- Field Data (from users actually accessing the site over weeks and months) and Lab Data (data from one machine at that given time) can produce wildly different scores.
- What we think of as ‘load speed’ encompasses thousands upon thousands of elements being loaded, many of which the user might never see. Most of the stuff that’s slow to load probably doesn’t impact the user’s experience.
Most importantly of all, there are a select few scenarios in which the load speed of a page will directly impact on that page’s ranking in search results
Core Web Vitals + Page Experience: The Gamechangers?
Let’s just talk about Core Web Vitals and Page Experience first. Google announced that Page Experience would be a ranking factor and began rolling out the update in the summer of 2021. Naturally, everyone in the SEO industry lost their shit and began to panic.
Core Web Vitals is essentially a set of metrics which show how ‘well’ a site loads. Page Experience takes these signals as well as mobile friendliness and a secure connection to decide whether the page delivers a good experience or not. If you don’t pass Page Experience, in theory your site’s visibility could suffer.
Previously, Google said publicly that Page Speed performance wasn’t a specific ranking factor, but that it might be used in a ‘tie-breaker’ situation to decide which page ranks above the other. In other words, if you had two sites of similar authority providing content of a similar quality, then the site which loads the fastest would probably win that head-to-head.
With the introduction of Page Experience though, that all changed, didn’t it?
Forget content and relevance. If you absolutely smash your Core Web Vitals, you’d see your SEMRush visibility bar climb steeper than a shilled Crypto shitcoin, right?
Hours and hours of client time would be spent trying to get a little circle to move from amber to green. Ecstatic SEOs posted their Lighthouse screenshots on Linkedin and waited for the likes to roll in.
Did anything really change, though?
Here’s a passage from Google’s official documentation on Page Experience:
While page experience is important, Google still seeks to rank pages with the best information overall, even if the page experience is subpar. Great page experience doesn’t override having great page content. However, in cases where there are many pages that may be similar in relevance, page experience can be much more important for visibility in Search.
Sounds very similar to the old way of using Page Speed as a ‘tie-breaker’.
I’m sure people cleverer than me have produced case studies and white papers on this already. But purely anecdotally, I’ve worked on improving the Core Web Vitals of a number of sites over the past year or so (sometimes reluctantly at the request of clients, sometimes because the site really was a mess) and found the gains to be very marginal.
Likewise, sites I work on that have awful Core Web Vitals have seen no noticeable fluctuations in rankings since the Page Experience update rolled out.
Again, totally anecdotal, but only by doing things yourself and seeing the results do you build a picture of what actually works, rather than relying on ’10 things to boost your site’s SEO’ blog posts.
When is Page Speed important?
A site’s performance can have a major impact on not just its SEO performance, but also the company’s bottom line. Huge ecommerce sites with huge revenue streams can suffer hugely if they provide a poor user experience which puts obstacles in the way of successful browsing and purchasing.
Poor user engagement signals as a result of a slow site can certainly impact on rankings, but only where a quick load is absolutely vital to that site’s functionality; when there’s lots of scrolling between different pages, when lots of media elements need to be loaded, or when lost of user interaction is required with the elements on the page.
Think shopping, news, entertainment, social – any website where browsing and flicking between pages quickly is vital functionality.
Does the user need the site to be lightning quick? Then site speed is probably important.
It’s probably very important. Working on improving it could bring thousands or even millions of pounds in additional revenue. In this case, you’d employ a technical SEO specialist to work exclusively on this one issue.
But most businesses investing in SEO in the UK don’t even come close to falling under this category. They pay agencies or consultants a flat fee to implement all SEO work, leaving up to them to decide how best they use their time.
When is Page Speed not important
Problems arise when those in the SEO industry push perfect Page Speed as the ultimate goal, the one silver bullet to kill all SEO demons (*cough* Neil Patel *cough*).
Marketing managers, business owners and junior SEO executives read these blogs and take their word as gospel. And while some of it may be valuable and helpful, the key to any worthwhile SEO campaign is identifying which implementations will bring the most positive impact and are worth spending time on.
For a huge multinational brand with a huge site with lots of technical issues, focusing on Page Speed would probably be worthwhile, if not absolutely essential.
For a small-to-medium sized businesses looking to invest in SEO to improve their visibility on a local or regional level, Page Speed would not usually be something that tops the list of priorities.
These sites tend to act like a brochure, a display of credibility and a source of more information on the company and their services. Some may have ecommerce functionality, but many don’t. Most are lacking basic optimisations around topical relevance and authority – two factors that Google will always value more than site performance.
Most sites I see in this category benefit most from some proper keyword research, a realignment of site structure to ensure that these keywords are being targeted properly, and ensuring that on-page fundamentals are in place to support that rejig. To move the needle further once these foundations are in place, I’d build backlinks and write supplementary pillar content to build topical authority around hub pages.
Only once all of this is done would I start to think about site speed as a priority, or at least only mop up some quick wins alongside the main work that wouldn’t eat too much into a client’s hours.
Why is this such a problem?
It’s a big problem when agencies and consultants burn large chunks of client retainer on trying to speed the site up a little bit, and giving themselves a big pat on the back when they do so, knowing full well it’s not likely to have a huge impact.
I’ve been guilty of this myself in the past through lack of experience, but any responsible SEO should be making sure they focus their hours on improvements they think will be most impactful. A lot of ‘technical’ tasks fall under this; disavowing backlinks is the biggest culprit.
It’s also a problem when business owners or marketing managers think that working on site speed would be a worthwhile endeavour and aren’t willing to listen to the advice of their SEO agency or consultant when they disagree, or think that they’re pulling the wool over their eyes by trying to avoid the task. They’ve read the top-ranking blogs and the YouTube videos; they know best.
The majority of businesses in the UK paying someone to implement SEO work on their site do not have a huge budget. The money they invest in their digital marketing needs to work for them – often it can make the difference between staying afloat or going under.
That’s why their hours need to be used wisely, on changes that will most likely bring results. Improving Page Speed usually isn’t one of them.
- Page Speed is important, but not as important as a lot of people in the industry make it out to be. At least not for the majority of sites paying for SEO work.
- By all means, your site should be as fast as possible. But a shit site that loads quickly is still shit.
- Work on improving site speed, if it’s an issue. But don’t prioritise it if it’s not one of the main factors holding the site back in search.
- Quality and relevance will always supersede site performance. Google dies the day that’s no longer the case.
- Good site performance can help a page to rank higher in certain scenarios, but usually the page already has to be the dog’s bollocks in order for Page Experience to come into the equation. If the page isn’t the most useful result for a certain search query, a perfect Page Speed score won’t make a blind bit of difference.
- There are usually many, many more impactful tasks to be implemented on a site before it’s worth spending time on improving Page Speed.