Core Web Vitals in Plain Language, and the 5 Fixes That Move Them

Posted September 24th, 2026 under Web development

Two colleagues discussing Core Web Vitals and website performance across a desk

Google’s own numbers say plenty about the state of the web: as of January 2026, only 55.7% of all web origins worldwide pass all three Core Web Vitals thresholds — the exact test Google uses to judge whether a page actually feels fast to a real visitor. If a site sits in that other 44%, people are noticing long before Google penalizes anything.

What do Core Web Vitals actually measure?

Core Web Vitals are three metrics Google pulls from real visits to real sites, not a lab test, measured at the 75th percentile across mobile and desktop. Largest Contentful Paint (LCP) tracks how long the biggest visible element takes to render. Interaction to Next Paint (INP) tracks how quickly the page responds the moment someone taps or clicks. Cumulative Layout Shift (CLS) tracks how much content jumps around while the page loads. Pass all three and Google counts the page as a good experience; miss one and it does not, however good the site looks in a screenshot.

LCP: how long until the main thing on the page actually shows up?

Good is 2.5 seconds or less; poor is anything over 4 seconds, measured at the 75th percentile. LCP usually gets dragged down by one of four things: a slow server response before the browser gets anything at all, CSS or JavaScript that blocks rendering while it loads, a hero image that is not discoverable early in the HTML, or an image that simply was not compressed. Vodafone cut LCP by 31% on its site and saw an 8% lift in sales, so the metric is not academic — it moves revenue.

INP: does the page respond the instant someone taps it?

Good is 200 milliseconds or less; poor is anything over 500. INP replaced First Input Delay as an official Core Web Vital in March 2024, and the difference matters: FID only ever measured the very first tap on a page, while INP watches every interaction for the whole visit. A page that opens fast but freezes when someone taps the menu, adds something to cart, or opens a form still fails INP, even if LCP looks perfect. The usual culprit is JavaScript: long tasks that block the main thread, or too much code running on every click.

CLS: does anything jump around while the page loads?

Good is a score of 0.1 or less; poor is anything past 0.25. The classic cause is an image or embed with no reserved space, so the layout shoves everything down the moment it finishes loading — exactly the moment someone was about to tap a button. Web fonts that render at a different size than the fallback font, and ads or widgets that resize themselves after the page has already settled, do the same thing.

Close-up of hands typing on a laptop keyboard while reviewing Core Web Vitals fixes

The 5 fixes that move all three metrics

Most sites do not need a rebuild to pass Core Web Vitals — they need five specific fixes, most of them a few minutes of work per template:

  • Give every image its dimensions. Set width and height, or aspect-ratio in CSS, on every image and embed so the browser reserves the space before the file loads. This is the single biggest CLS fix on most sites.
  • Tell the browser which image matters most. The hero image should load first, not compete with a dozen other requests. Mark it high priority and skip lazy-loading on it:
<img src="hero.jpg" alt="…" fetchpriority="high" width="1200" height="900">
<link rel="preload" as="image" href="hero.jpg">
  • Trim and defer JavaScript that is not needed immediately. Every script that runs before someone can interact with the page adds to both LCP and INP.
  • Self-host fonts and set font-display: swap. A font pulled from a third-party server on every visit is one more render-blocking request; hosting it locally and letting the fallback font display immediately prevents both a layout jump and a render delay.
  • Move to hosting built for speed, not just uptime. Time to First Byte is the floor every other LCP fix builds on — no amount of image optimization fixes a server that takes two seconds to respond before the page even starts loading.

What this means for a small Aruba business

Mobile pass rates for Core Web Vitals sit lower than desktop across the web — 49.7% versus 57.1%, according to the HTTP Archive’s CrUX data — and that gap matters more here than most places. A large share of the traffic hitting an Aruba business site is a visitor on hotel wifi or a phone still roaming, not someone on a fast home connection. A multilingual site running English, Spanish, Dutch and Papiamento versions carries more markup and more font weights than a single-language site, which raises the stakes on every one of the five fixes above rather than lowering them. 53% of mobile visitors abandon a page that takes more than three seconds to load, and on an island where most first contact with a business happens through a phone in someone’s hand, that is not a rounding error.

Related: see how WordPress maintenance habits keep small regressions like these from creeping back in after every plugin update.

Hand writing a Core Web Vitals checklist by hand on paper

A checklist to run this week

  • Check your real numbers first. Google Search Console’s Core Web Vitals report and PageSpeed Insights both show the actual 75th-percentile scores for your site, not a guess.
  • Add width and height to every image tag on the homepage and the two or three pages that get the most traffic.
  • Find the render-blocking scripts loading before the visible content and defer anything that is not essential.
  • Test on an actual phone over mobile data, not just a laptop on office wifi, since that is the connection most visitors are really using.

We handle speed and Core Web Vitals as part of our WordPress maintenance plans for businesses in Aruba.

Talk to 139 Design Studio about a real performance review — image handling, script cleanup, fonts and hosting, plus the maintenance plan that keeps a fast site fast after the next plugin update.

Sources: web.dev — Largest Contentful Paint (LCP) · web.dev — Interaction to Next Paint (INP) · web.dev — Cumulative Layout Shift (CLS) · SHNO — Core Web Vitals Statistics 2026