Speed is an important ranking factor, and many optimizations can be made to improve it – but having a deep understanding of exactly which metrics to improve is what gives webmasters and SEOs a competitive edge.
Back in July 2018, Google shared some data on the aftermath of the release of the ‘Speed Update’, which reduced the rankings of very slow mobile pages in the search results.
The results were positive. 95% of countries had improved speeds, and there was a 20% reduction in abandonment rate for navigations initiated from the search results.
But perhaps most importantly of all, user-centric performance metrics improved by 15% to 20% for the slowest third of traffic.
When webmasters and SEOs think about page speed performance, they can get caught up in focusing solely on trying to achieve a fast full load time for a given URL. This is a common error that can often result in a reel of initiatives devised to reduce the amount of time it takes for a page to fully load.
When reporting on speed performance, it’s important to be precise so that we, as SEOs, can lead developers to optimize for the right things. Otherwise, we risk harming the user experience, rather than improving it.
Whether a user perceives a page to load fast is governed by multiple moments during the load experience. So, as you try to answer the question: ‘How fast is my website?’ You’ll realize that the term ‘fast’ is vague. What exactly does Google mean when they say fast? In what context? And for whom?
Let’s take another look at the page speed results reported by Google. Notice how they didn’t report anything on full load time, but they did report on user-centric performance metrics? Why? Well, it’s because these are the key performance indicators that directly affect the experience of your users – not full load time.
The key takeaway? Optimize your site to deliver an exceptional user experience, rather than simply a super fast full load time, and you will be rewarded by Google.
So, what are the magic metrics that you should optimize for?
When a user searches in Google and navigates to a web page, they expect to see visual feedback and be able to interact with the elements to reassure them that the page is loading. With this in mind, there are three key user-centric performance metrics worth paying attention to:
|The Experience||The Metric|
|Is it happening?||First Paint (FP) / First Contentful Paint (FCP)|
|Is it useful?||First Meaningful Paint (FMP) / Hero Element Timing|
|Is it usable?||Time to Interactive (TTI)|
‘Is it happening?’ is the first critical moment in a page’s loading path that can affect a user’s load perception.
In simple terms, First Paint (FP) and First Contentful Paint (FCP) are the metrics which pinpoint when the browser renders the first pixels on the screen. This happens immediately after a user has navigated to a page from the search results.
The First Paint metric indicates when the browser renders the first visually different screen from the navigation, whereas First Contentful Paint is when the browser renders the first bit of content, such as a piece of text or an image.
These two metrics are the first port of call when assessing whether you are delivering a fast page speed performance.
As a guideline, aim for FP / FCP to load in under 2 seconds on a fast 3G connection.
‘Is it useful?’ is the next moment in the load path that you should pay attention to.
As opposed to FP / FCP, where the browser renders the first pixels and piece of content, First Meaningful Paint (FMP) / Hero Element Timing refers to when the most important parts of a web page load.
An example that Google gives is to imagine you click on a YouTube result – the most important part of that webpage for you is the video. Or, on Twitter, for example – it’s the notification badges, and so on.
Knowing which elements of your page are most important will help your web developers to optimize the pages to load the parts that are going to be most useful for your users as quickly as possible.
Ultimately, FMP measures when a user perceives that the primary content of a page is visible. It’s measured as the time between the user clicking on a search result and the page loading and rendering its primary content above the fold.
A guideline to follow here would be to aim for FMP to load in under 3 seconds on a fast 3G connection.
‘Is it usable?’ is the third moment in the load path which is measured using Time to Interactive (TTI). It pinpoints the time at which your web page has visually rendered and – most importantly – is ready to respond to user input.
Serving a page that looks ready but doesn’t appear to work can easily lead users to get frustrated and, potentially, to bounce back to the SERPs. If you consider the theory of the long click to be valid, this in turn can lead to reduced rankings.
TTI can, therefore, be considered as the most important user-centric performance metric to get right. Imagine you optimize your site to load FP, FCP and FMP in under 3 seconds on a 3G fast connection, but the user is then forced to wait a long time for the pages to become interactive?All the work you’ve done to improve your load path for the user could be wasted.
To take advantage of the optimizations you’ve made to get FP, FCP and FMP all to load in under 3 seconds, aim for TTI to mark the 5 second point in the page load time on a 3G fast connection.
As with everything in SEO, in theory, you should be able to achieve higher rankings by optimizing for FCP, FMP and TTI, especially if your competitors are pouring their efforts into improving full page load time. It is well documented that Google use these metrics to determine page speed performance, so optimizing for them is the smartest move you can make.
Aside from improved rankings and increased traffic, another major benefit to being precise when reporting on speed performance is the ability to use metric-specific data to spot how performance affects business.
Let’s take the example of lead generation websites. Imagine you have a lead generation website which collects leads through form completions, and you have goal completion tracking set up for the completion of a form. You make improvements to your FP and FCP but see no improvement in conversion rate. You further your efforts to improve TTI and you see form completions increase and as such, conversion rate improves.
Here you can make the assessment that faster interactive times lead to increased form completions, and result in improved business performance.
From the example above it is easy to see why you should focus on user-centric metrics when it comes to delivering fast experiences, as opposed to optimizing for a generic full page load time.
What you need to optimize to achieve optimal performance is totally dependent on the makeup of your website.
Google have outlined how webmasters with full control of their websites can optimize performance and preventing regression for all three of the user-centric metrics discussed.
Given that not all of us are technically minded, a lot of small businesses now turning to the simpler option of website builders. If your website was built using Wix, Squarespace, Shopify or any of the other online builders, you’ll have limited control over what you can optimize. For example, you can ensure that images are compressed and scaled to size, but the website builder will determine the technical layout of your site.
Josh Frisby is a passionate SEO enthusiast who launched his career building his own lead generation websites before creating and delivering SEO strategies for leading digital marketing agencies, multinational market leaders and the UK’s fastest growing tech firm. Josh has SEO expertise spanning across lead generation to affiliate marketing and ecommerce. Learn more about Josh Frisby on LinkedIn and be sure to visit Josh’s website: Website Builder Expert.