How to Fix Cumulative Layout Shift (CLS) in 2026
You go to tap a button and the page jumps, so you tap an ad instead. That's Cumulative Layout Shift, and Google measures it as a ranking signal. Here's exactly what causes it and how to drive your CLS score down to where it should be.

You're reading a paragraph, an image loads above it, and the text lurches down the screen. Or worse, you go to tap a link and an ad slot pops in at the last instant, so your finger lands somewhere you never meant to go. That flicker of frustration has a name and a number attached to it.
Cumulative Layout Shift measures how much your page moves around while it loads, and Google treats it as part of the page experience it uses to rank you. It's one of the three Core Web Vitals, and unlike raw speed, it's often caused by a handful of specific, fixable mistakes. The good news is that CLS is one of the most solvable problems in technical SEO once you know where the jumps come from.
Quick Answer: To fix Cumulative Layout Shift in 2026, reserve space for anything that loads late so nothing shoves the rest of the page around. Set explicit width and height on images and video, give ad and embed slots a fixed container, preload fonts and set size-adjust so text doesn't reflow, and never inject banners or content above what a user is already reading. A CLS score of 0.1 or below is the target, and most sites reach it by eliminating a few specific shift sources.
- CLS measures unexpected movement of page elements while content loads
- A good score is 0.1 or below, and above 0.25 is rated poor
- The top causes are unsized images, ad slots, late fonts, and injected content
- Setting width and height on media reserves space and stops the biggest shifts
- Field data from real users is what Google ranks on, not lab tests alone
- Fonts cause shifts when fallback text reflows, which preloading helps prevent
- Never insert banners or notices above content a user is already reading
What Actually Causes a High CLS Score?
A short list of usual suspects, almost every time. CLS isn't random, it's the sum of elements that load without a reserved place and then push everything around them when they arrive. Once you know the patterns, you can spot most of them in a single scroll through your own page.
The biggest offender is media without dimensions. When an image or video has no declared width and height, the browser doesn't know how much space to hold for it, so it renders the surrounding content first and then jolts it down when the media finally loads. Ads and third-party embeds do the same thing on a larger scale, dropping into slots that had zero reserved height. Web fonts add another layer, since a fallback font shows first and the real font reflows the text once it downloads. And then there's the self-inflicted kind, banners, cookie notices, and dynamically inserted content that appear above whatever the reader is looking at.
Here's how the common causes map to their fixes at a glance.
| Cause of shift | Why it happens | The fix |
|---|---|---|
| Images without dimensions | Browser reserves no space | Set explicit width and height |
| Ads and embeds | Slots have no reserved height | Give each slot a fixed container size |
| Web fonts | Fallback reflows to real font | Preload fonts, use size-adjust |
| Injected banners | Content pushed down after load | Reserve space or place below the fold |
| Late dynamic content | Elements added above viewport | Insert below current scroll position |
The insight worth holding onto is that CLS is almost always about space, not speed. A page can load quickly and still shift badly, because the problem isn't how fast things arrive, it's whether the browser knew where to put them before they did. Fix the reservations and the score drops.
How Do You Measure CLS Accurately?
With two kinds of data, and you need both to trust what you're seeing. Lab data comes from a controlled test of a single page load, while field data comes from what real visitors actually experienced. Google ranks on the field data, so that's the one that ultimately counts, but lab tools are how you find and fix the specific shifts.
Field data lives in the Core Web Vitals report inside Search Console and in the Chrome User Experience Report, which aggregates real Chrome users over a rolling window. This is the source of truth for how Google sees you, and it's grouped by URL patterns so you can find which templates are failing. The catch is that it lags, since it reflects the last 28 days of visits, so a fix you shipped today won't fully show for weeks. Don't panic when the number doesn't move overnight.
Lab tools are for diagnosis and instant feedback. Lighthouse, PageSpeed Insights, and Chrome DevTools all report a CLS figure for a single load and, crucially, let you see which elements shifted. DevTools has a performance panel that highlights the exact nodes that moved, which turns a vague "your CLS is high" into a precise "this image and this ad slot are the problem." One important quirk to remember, CLS accumulates over the full life of the page including interactions and scrolling, so a lab test that only measures the initial load can understate what real users hit further down. Test the way people actually use the page, scrolling and interacting, not just the first paint.
How Do You Fix Layout Shifts From Images and Ads?
You reserve the space before the content shows up, full stop. Almost every image and ad shift disappears the moment the browser knows how tall the element will be, because then it holds that room open and slots the content in without moving anything.
For images and video, set explicit width and height attributes on every element, or use the CSS aspect-ratio property so the browser can compute the space from the ratio alone. Modern browsers use those dimensions to reserve the correct area even before the file downloads, so the surrounding text never jumps. Combine that with responsive sizing so the image scales without losing its reserved proportions, and one of the largest sources of CLS on most sites just vanishes. This pairs naturally with proper image SEO, where dimensions, format, and lazy loading all work together.
Ads and third-party embeds need the same treatment with a twist, since their content is out of your control. Wrap every ad slot in a container with a fixed minimum height matched to the ad size you expect, so the space is claimed whether or not an ad fills it. If a slot can serve different sizes, reserve the largest common one to avoid a collapse-and-jump when a smaller ad loads. The same logic applies to embedded videos, maps, and social widgets, which should each sit inside a sized container rather than dropping in raw. It costs a little vertical space when a slot goes unfilled, and that tradeoff is almost always worth a stable page.
How Do You Stop Fonts and Injected Content From Shifting?
You make text render in its final position the first time, and you stop putting things above where people are reading. These two causes are sneakier than images because they feel unavoidable, but both have clean solutions.
Fonts shift because the browser shows a fallback typeface first, then swaps in your web font once it loads, and the two rarely occupy the same space. The fix is a combination. Preload your critical font files so they arrive early, use font-display settings that avoid invisible or reflowing text, and apply the size-adjust and metric override properties so your fallback font takes up nearly the same room as the real one. When the fallback and the final font match in size, the swap becomes invisible and the text never jumps. Self-hosting your fonts rather than pulling them from a third party also removes a slow network hop that widens the reflow window.
Injected content is the one entirely within your power, so there's no excuse for it. Cookie banners, promo bars, notification strips, and any content added by script after load should never appear above what the user is already viewing, because doing so shoves everything down at the worst possible moment. Reserve space for these elements from the start, or place them at the bottom, or overlay them so they float above the page without displacing it. The same rule covers anything your JavaScript inserts dynamically, which should always render below the current scroll position or into a pre-sized slot. Respect where the reader's attention is, and you eliminate the most jarring shifts of all.
FAQ
What is a good CLS score in 2026? The threshold hasn't changed, and 0.1 or below is considered good. Between 0.1 and 0.25 needs improvement, and above 0.25 is poor. Google evaluates the score your real users experience, measured at the 75th percentile, so you want most visits landing under 0.1. It's an unusual metric in that lower is always better and the ceiling for "good" is quite strict.
Does CLS really affect my rankings? It's part of the page experience signals Google uses, so it's a real factor, though a modest one compared to relevance and content quality. It rarely makes or breaks a page on its own, but it can be a tiebreaker between similar results, and a badly shifting page also drives users away through sheer frustration. Fixing it helps both your rankings and your actual visitors.
Why does my lab CLS look fine but field data is poor? Because lab tools usually measure only the initial page load, while CLS accumulates across the whole session including scrolling and interactions. Real users trigger shifts your single-load test never sees, like lazy-loaded images further down or content that appears after they interact. Trust the field data as the real picture, and use lab tools with scrolling and interaction to reproduce what users hit.
How long until my CLS fixes show up in Search Console? Expect weeks, not days. The Core Web Vitals report is built on field data covering roughly the last 28 days of real visits, so a fix needs time to accumulate enough new sessions to move the average. Lab tools confirm your fix works immediately, but the official field number lags. Ship the fix, verify it in the lab, and be patient with the report.
Do lazy-loaded images cause CLS? They can, if they load without reserved dimensions, because the space isn't held until they appear further down the page. Lazy loading itself is fine and good for performance, the trick is to always set width and height or an aspect ratio on lazy-loaded images too. With dimensions declared, the browser reserves the space up front, so the image fades in without shoving anything around.
Can animations trigger layout shift? Animations that change an element's size or position in the document flow can, yes. The fix is to animate using CSS transform and opacity instead of properties like top, height, or margin, because transforms move an element visually without reflowing everything around it. Well-built animations that stick to transform and opacity don't count against your CLS at all.
Where to Go Next
Start with the biggest lever, which is dimensions. Add explicit width and height or an aspect ratio to every image and video, wrap every ad and embed in a sized container, and you'll knock out most of your CLS in an afternoon. Then handle fonts with preloading and size-adjust, and audit anything your scripts inject above the fold.
Verify as you go with lab tools for instant feedback, then watch the field data in Search Console settle over the following weeks. CLS is the most fixable of the three Core Web Vitals, so once it's clean, turn to the others with the guides on fixing Largest Contentful Paint and optimizing INP. Astro SEO Blog covers Core Web Vitals and page experience across the technical SEO category, so you can get all three into the green.
Related Articles

How to Optimize for Bing and Microsoft Copilot in 2026
Almost every SEO guide obsesses over Google and forgets Bing exists. That's a mistake in 2026, because Bing quietly powers Microsoft Copilot, ChatGPT search, and a slice of traffic you're leaving on the table. Here's how to win it.

How to Control AI Crawlers With robots.txt and llms.txt in 2026
GPTBot, ClaudeBot, and Google-Extended are hitting your site right now. Here's how to decide who gets in, block the ones you don't want with robots.txt, and whether the new llms.txt file is worth your time.

How to Earn Sitelinks in Google Search
Sitelinks are the extra links Google adds under your listing, and you can't add them manually. Here's how site structure, internal linking, and clear navigation make Google generate them, and why you can only influence, not control, them.