What Are Core Web Vitals? A Practical Guide to Site Speed and UX (2026)

Site speed is the most talked-about and most mismeasured topic in SEO. What Google looks at is not the score a speed testing tool gives you; it is three metrics that measure the experience real users have on the page. These three are called Core Web Vitals, and knowing what they are means knowing what to fix.

Three metrics, three different questions

LCP — when did the main content appear?

Largest Contentful Paint measures how long it takes for the largest image or text block on the page to appear on screen. From the user’s point of view: when did the page load? The element being measured is usually the hero image, a product photo or the headline block. The threshold considered good is under 2.5 seconds.

INP — when I clicked, how fast did it respond?

Interaction to Next Paint measures how quickly the page responds visually to user interactions, and it has replaced the older FID metric. You tapped the menu, picked a filter, hit “add to cart” — how long did it take for something to change on screen? The good threshold is under 200 milliseconds.

CLS — did the page shift under me?

Cumulative Layout Shift measures how much elements move around while the page loads. Text jumping down while you are reading, or a button moving just as you are about to tap it, hurts this metric. The good threshold is below 0.1.

Lab data vs. field data

The score you get from speed tools is lab data: it is generated by simulating a standard device and connection. What Google uses as a ranking signal is field data, collected from real users on real devices. That is why a site that scores high in a testing tool can still look weak in field data. Base your decisions on field data; lab data shows you what to fix, not where you stand.

What hurts these metrics the most?

• Unoptimized images. By far the number one cause of LCP problems: images that are not properly sized, not converted to modern formats and not prioritized.
• Page builder plugins and bloated themes. The CSS and JavaScript for dozens of features you never use gets downloaded by every single visitor.
• Render-blocking resources. Files queued up in the head section delay the first paint.
• Third-party scripts. Chat widgets, heatmaps, multiple analytics tools and pixel codes directly hurt INP.
• Images and ad slots without defined dimensions. The primary cause of CLS: when no placeholder is defined, the page jumps around as content loads.
• Late-loading web fonts. Font swapping causes both layout shift and visual instability.

Order of fixes: start with the biggest wins

1. Fix your images. Correct dimensions, modern formats, a priority hint for the above-the-fold image and lazy loading for everything further down. On most sites, this step alone brings LCP below the threshold.
2. Remove unnecessary plugins. Every plugin comes at a cost. Unused plugins should be deleted, not just deactivated.
3. Count your third-party scripts. How many of them do you actually use? Every unused pixel is pure dead weight.
4. Load critical CSS first, defer the rest. Anything beyond the styles needed to render the first screen should wait.
5. Define placeholders. Give images, iframes and ad slots an aspect ratio.
6. Fix caching and the server side. If server response time is high, every other optimization will only have a limited impact.

In our own projects we don’t use page builder plugins; we code the theme directly — not for aesthetic reasons, but precisely because of this cost. We explain the approach in our corporate web design guide.

How speed affects conversions

Before it is an SEO metric, speed is a commercial metric. A product page that loads slowly on mobile lowers your add-to-cart rate; a slow form page gets abandoned. If you run ads, the effect doubles: you get fewer conversions for the same click budget. That is why checking your landing page speed before increasing your ad budget is almost always the more profitable move; we cover the budgeting logic in our Google Ads budget article.

3D and animated websites

Sites with motion and 3D scenes can easily hurt these metrics, but they don’t have to. Making sure the scene doesn’t hold up the first screen, doesn’t render before it enters the viewport and uses a poster frame keeps all three metrics healthy. You’ll find the details in our article on Core Web Vitals for 3D websites.

Frequently asked questions

Are Core Web Vitals a ranking factor?

Yes, but not a decisive one on their own. Content relevance is a stronger signal. In practice, the impact shows up like this: of two pages of similar strength, the one with the better experience comes out ahead. In other words, speed alone won’t win you rankings, but it can lose them.

Does my speed score need to be 100?

No. The goal is a threshold, not a score. It is enough for all three metrics to be in the good range. The final effort spent pushing the score up to 100 usually has no commercial return.

Is WordPress a slow platform?

Not by itself. The cause of slowness is almost always the choice of theme and plugins. WordPress sites built with discipline pass all three metrics comfortably.

Is installing a caching plugin enough?

It helps partially, but it doesn’t fix the root cause. Caching hides the load a bloated theme generates; it doesn’t eliminate it. Its impact on INP in particular is limited.

How long does it take to see improvement?

In lab data, the effect is immediate. Field data, however, is based on a rolling window of roughly the past month, so it takes a few weeks for improvements to show up in reports. If you’d like us to assess your site’s speed, get in touch.