Core Web Vitals Explained for Business Owners: What Google Measures and Why It Matters

What Google introduced and why most websites still fail it

In 2021, Google formalized something it had been building toward for years: a set of measurable, technical page experience signals that would directly influence search rankings. These became known as Core Web Vitals, three specific measurements of how a page feels to use, not just how fast it technically loads.

The timing matters. Google announced Core Web Vitals in 2020 with a year’s notice before they became ranking signals. The official documentation on web.dev was updated months in advance. The developer community was aware. Most business owners were not.

Four years later, HTTP Archive data shows that a significant portion of websites still fail at least one Core Web Vital metric on mobile. The gap between “site owners who know their score” and “site owners whose score is good” is substantial. The consequences of that gap are measurable in search rankings and conversion rates.

The three measurements and what they actually mean

LCP, Largest Contentful Paint measures how quickly the main content of a page becomes visible to the user. In practice, for most business websites, this is the hero image or the main headline. The benchmark is 2.5 seconds or under. A site loading its hero in 4 seconds has a Poor LCP score.

Why does this matter beyond rankings? Nielsen Norman Group’s research on user attention shows that users form first impressions of a page within 50 milliseconds and begin abandoning pages that feel slow within the first 2-3 seconds. A page with Poor LCP is losing a measurable percentage of visitors before they see any content. The business didn’t fail to attract them, it failed to retain them long enough for them to decide to stay.

The most common LCP culprit is the hero image. A full-width background image at 2.5MB, served as JPEG without compression, from a shared hosting server with slow response times, causes LCP failures on nearly every site that uses this configuration. Converting the image to WebP format, reducing its file size to under 150KB, and adding a loading="eager" and fetchpriority="high" attribute tells the browser to prioritise loading it, these changes typically move LCP from 4+ seconds to under 1.5 seconds. This is not a redesign. It is a configuration change.

CLS, Cumulative Layout Shift measures how much the visible content of a page moves around while it’s loading. A CLS score of 0.1 or under is Good. A high CLS score means text shifts, buttons move, and users accidentally click the wrong thing because the layout wasn’t stable when they tried to interact with it.

The most common causes of high CLS: images without defined width and height attributes cause the browser to not reserve space for them, so when the image loads, everything below it jumps down. Ads and cookie consent banners that load after the initial content push everything else down. Fonts that load late cause text to reflow when the web font replaces the fallback font.

All of these are fixable without visual changes to the design. The layout looks identical after the fix. The behaviour during loading is completely different.

INP, Interaction to Next Paint replaced FID (First Input Delay) as a Core Web Vital in March 2024. It measures how quickly the page responds when a user clicks a button, taps a link, or interacts with a form. Under 200ms is Good. A page that takes 500ms to respond to a button click feels broken, even if technically the button works.

INP failures are almost always caused by JavaScript execution. Heavy JavaScript bundles that run synchronously block the browser’s ability to respond to user interaction. The fix is code splitting (loading JavaScript only when needed) and moving non-critical scripts to load after the page is interactive.

The business case that isn’t about rankings

Rankings are one consequence of Core Web Vitals scores. The conversion rate impact is often more significant for the business’s direct bottom line.

Google’s own research on the relationship between performance and conversion shows that a 0.1-second improvement in load time increases retail conversion rates by 8.4% on average. Portent’s study of conversion rates by load time shows a site loading in 1 second converting 3x better than a site loading in 5 seconds. These numbers are consistent across multiple independent research bodies, not because of SEO, but because a fast page is a better user experience.

A business with an e-commerce site generating €50,000 per month in revenue, running with Poor Core Web Vitals, and improving to Good scores can realistically expect a 15-25% increase in conversion rate from the performance improvement alone, not from any change in the product, pricing, or marketing. The visitor pool is the same. The conversion rate is different because the experience is different.

For B2B service businesses and e-commerce brands, this is not a technical argument. It is a revenue argument.

How to check where you stand right now

Google PageSpeed Insights is free and takes two minutes. Enter any URL, run the mobile test, and look for three things:

The LCP score and the element Google identifies as the Largest Contentful Paint. If it’s a large image, image optimization is likely the primary fix. If it’s a text block, server response time and render-blocking resources are the likely culprits.

The CLS score and the specific elements Google flags as layout shifts. These are often fixable in under an hour once identified.

The INP score and the specific JavaScript tasks contributing to slow interaction response. These typically require a developer to address but are often caused by specific third-party scripts rather than the site’s own code.

At x078, the Core Web Vitals baseline is set before any visual design begins on a new landing page or website. The technical architecture (static output, optimized images, deferred scripts) is chosen specifically to produce Good scores by default, not as an afterthought. A site that requires retrofitting to pass Core Web Vitals after launch costs more to fix than a site built to pass them from the start.

The score is measurable. The financial impact is calculable. The fix is known. The question is whether it gets done before or after the revenue impact accumulates.