Imagine opening a service page on your phone. The heading takes ages to appear. You tap “Get a quote” and nothing seems to happen. Just as you tap again, an image loads and moves the button. That is the experience Core Web Vitals aim to describe.

The short answer: Google's current Core Web Vitals are Largest Contentful Paint (LCP) for loading, Interaction to Next Paint (INP) for responsiveness and Cumulative Layout Shift (CLS) for visual stability. They help you find problems real visitors can feel. They are useful signals, but a fast page still needs helpful content and a clear route to the next step.

What does each Core Web Vital measure?

Metric Plain-English question Google's “good” threshold
Largest Contentful Paint (LCP) How quickly does the main visible content appear? 2.5 seconds or less
Interaction to Next Paint (INP) How promptly does the page respond when someone uses it? 200 milliseconds or less
Cumulative Layout Shift (CLS) How much does visible content unexpectedly move? 0.1 or less

Google evaluates these targets at the 75th percentile of page loads, separately for mobile and desktop. That means the experience needs to be good for most visits, not just on a fast office connection. The thresholds and current metric set are described in Google's Web Vitals guide.

LCP: the main content appears too late

LCP often reflects a large hero image, banner or block of text. A high-resolution image uploaded without suitable sizing can make the main content slow to appear. Other causes include slow server responses, render-blocking resources and loading a critical image later than necessary.

Resize and compress images appropriately, serve modern formats where supported and inspect whether the main image is being prioritised. Good hosting and a sensible cache can help, but the right fix depends on the actual bottleneck.

INP: the page reacts slowly

INP looks at responsiveness across a visitor's interactions. A menu may open late, an add-to-basket button may hesitate or a form may seem frozen. Heavy JavaScript, busy third-party widgets and long tasks in the browser can contribute.

Test the actions people genuinely take: open navigation, select a product option, add an item to the basket and submit a form. Removing an unused script or changing how work is handled may have more impact than another caching setting. A simulated page load without interaction cannot directly measure INP; lab tools may use other diagnostics as clues.

CLS: the page jumps around

CLS captures unexpected visual movement. Images without reserved space, a banner inserted above the content or fonts swapping late can push text and buttons around. The obvious symptom is trying to tap one thing and hitting another because the page moved.

Set image and embed dimensions, reserve space for content that loads later and avoid injecting elements above what a visitor is reading. Some movement following a deliberate user action is expected; the concern is movement that surprises people.

Why do PageSpeed scores and real-user results differ?

PageSpeed Insights may show field data, based on real Chrome users over a recent period where enough data exists, and lab data, a controlled test used to diagnose potential causes. A fresh test can improve before the field figures reflect the change. Some pages have too little real-user data to show a page-level result.

Use field data to see what visitors have experienced and lab diagnostics to investigate why. Test the important page types separately: a homepage, a service page and a product page may have entirely different problems. Read the mobile results carefully and confirm a fix by testing the actual interaction, not just watching a single score rise. Google's Core Web Vitals tools guide explains the different reports.

Do Core Web Vitals affect SEO?

Google includes Core Web Vitals in its wider consideration of page experience, but passing the thresholds does not guarantee higher rankings or more sales. Relevant content and a page that answers the searcher's need remain essential. Performance work is valuable because it makes the site easier to use as well as addressing a technical quality signal.

For a business website, the sensible order is to find the pages people use to enquire or buy, identify the metric causing trouble and fix the underlying issue. Then check that the journey still works: forms submit, menus open and checkout remains reliable. A 100/100 lab score is not a business goal in itself.

Where should you start?

Run your busiest pages through PageSpeed Insights and check the Core Web Vitals report in Google Search Console. Group similar pages with the same issue, such as oversized hero images across service pages. Prioritise the change that improves the experience of the most relevant visitors, then monitor the result over time.

If you are unsure what a report is telling you, Vannquish Creative can translate the measurements into a practical list of fixes and check that the website feels better to use after the work is done.