Making sure your site is fast and responsive is important for your visitors — and because of this, Google also factors site speed into search rankings. Speed is not and has never been a “major” ranking factor (and Google’s John Mueller called the addition of Core Web Vitals “quite a subtle update”), but it is nevertheless important if you want to stay competitive.
In June 2021, Google added Core Web Vitals to the Page Experience algorithm for mobile (and in February 2022 for desktop). These are three metrics (and will likely be more in the future) designed to give a sense of how fast your site is, how quickly it responds to user input, and how much items shift around on the page. Collectively, the idea is to ensure that you are providing a good and fast experience for your visitors.
Because speed is one of the few ranking factors that Google actually tells us about, people tend to overestimate how important it is for SEO. It’s really just part of hundreds of factors Google considers — and it’s definitely not as important as having great content that provides value for your readers. In other words, site speed and Core Web Vitals are not “make or break” for a site — as long as a site is not extremely slow.
The page experience update introduces a new signal that our search algorithms will use alongside hundreds of other signals to determine the best content to show in response to a query. Our systems will continue to prioritize pages with the best information overall, even if some aspects of page experience are subpar. A good page experience doesn’t override having great, relevant content.Google’s FAQ on Core Web VItals
Core Web Vitals also were rolled out for mobile in mid-June 2021. This was sandwiched between two big “core” updates in early June and early July. The core updates caused a lot of volatility, but since it was less “hyped” than the Core Web Vitals rollout, many people attributing ranking changes to Core Web Vitals and site speed, when it really was the core updates instead.
The three Core Web Vitals metrics are:
- Largest Contentful Paint (LCP) is the largest element on the page in the viewport when the page is first loaded, before scrolling (this is usually the top of the page, “above the fold”). The idea is that the largest thing visible is important, so this is a stand-in metric for how quickly the page loads for the visitor. It’s often the first image in a post, or your header image, a blog post title, or maybe the first paragraph of text — and that can vary from page to page.
- Cumulative Layout Shift (CLS) is a measurement of how much things move around on the page, both during the initial page load and afterwards while the visitor is using the site. The idea is to avoid annoying visitors by having things move around on the page after they’re visible. It looks at how big each item is, and how far it moves, to calculate the score. It’s “cumulative” since things can continue moving around as the visitor is on the page.
How are they measured?
Google collects real-world timing data (“field data”) for the Core Web Vitals as people use the Chrome web browser. This is turned into the the Chrome User Experience Report (CRuX), and this database is what Google uses for its ranking algorithm.
It takes up to 28 days to collect this data, so if you make changes, these numbers will update slowly during the following weeks. You can see the data by running a test in Google’s PageSpeed Insights testing tool or by viewing the Page Experience report in Google Search Console for your site.
Google collects data on individual URLs. You can also see an “Origin Summary” which shows a summary of the results for all URLs on your domain. A PageSpeed Insights report will look something like this:
How good does a site need to be?
Google gives a goal for each metric separately, and breaks each up into “Good,” “Needs Improvement,” or “Poor.”
Since the times will be a little different every time someone loads a page on your site, they look at the 75th percentile of page loads and use that as the score. So if 75% of pageviews have a LCP time of 2 seconds, that’s your overall score. This allows for pages to sometimes load a little bit slower and that’s okay.
For Largest Contentful Paint, 2.5 seconds or lower is considered “good.” (2.5 to 4 seconds is “needs improvement,” and 4 seconds or slower is “poor.”)
For First Input Delay, 100ms or lower is “good.” (100ms to 300ms is “needs improvement, and 300ms or slower is “poor.”)
For Cumulative Layout Shift, a score of 0.10 or lower is “good.” (0.10 to 0.25 is “needs improvement,” and 0.25 or higher is “poor.” )
When Core Web Vitals were initially announced, they were “all or nothing,” meaning you’d need to “pass” (or be in the “good” range) for all three scores. However, in May of 2021, Google relented on this position, and now it is not all-or-nothing. The closer you are to the “good” range for each metric, the better for SEO. And then, once you’re in the “good” range, there’s no additional SEO benefit for further improvement — Google considers the “good” threshold to be a “very high bar.” Note, though, that further improvements may still be better for your readers.
Frustratingly, Google continues to use “Pass/Fail” language in the PageSpeed Insights testing tool and in Google Search Console. However, Core Web Vitals is definitely not pass/fail.
Realistically, this means that if your LCP is, for example, 2.6 seconds and is therefore flagged as “needs improvement” the actual real-world benefit of getting to 2.5 seconds (“good”) is going to be negligible.
Where to Test Your Site
Andrew discusses Core Web Vitals on the Food Blogger Pro Podcast.