Why Page Speed Became a Business Problem
In 2021, Google made page speed a confirmed ranking factor through its Core Web Vitals update. In 2024, INP (Interaction to Next Paint) replaced FID as the responsiveness metric, raising the bar further. In 2026, the sites that have invested in performance are quietly outranking and outconverting those that haven't.
But rankings aside, the conversion case for page speed is straightforward: every 100ms increase in page load time reduces conversion rate by approximately 1%. For a site doing £500,000 in annual revenue, a 1-second load time improvement is worth roughly £5,000–£25,000/year in additional revenue.
This article explains what Core Web Vitals are, why they matter, and what to actually fix.
The Three Core Web Vitals (2026)
LCP — Largest Contentful Paint
What it measures: How long until the largest visible element (usually a hero image or heading) renders on screen. Good: ≤2.5 seconds | Needs work: 2.5–4s | Poor: >4s
This is the metric most UK websites fail. The most common cause: large unoptimised hero images loaded at full resolution with no lazy loading, compression, or modern format (WebP/AVIF).
Fixes:
- -Compress and convert hero images to WebP or AVIF (typically 30–70% smaller than JPEG/PNG)
- -Preload the LCP image (add a preload link tag to your HTML head with rel=preload and as=image)
- -Use a CDN to serve images from a location close to UK visitors
- -Lazy load everything below the fold, eager load everything above it
CLS — Cumulative Layout Shift
What it measures: How much the page visually shifts around as it loads. The metric UK developers hate most because it's caused by subtle mistakes. Good: ≤0.1 | Needs work: 0.1–0.25 | Poor: >0.25
Common UK website causes: images without explicit width/height attributes, web fonts loading and causing text reflow, ads or embeds that inject without reserved space.
Fixes:
- -Always specify width and height on image elements
- -Use font-display: optional or font-display: swap for web fonts, with explicit font metrics
- -Reserve space for dynamic content (ads, iframes) with explicit dimensions
- -Avoid injecting content above existing content after load
INP — Interaction to Next Paint
What it measures: How long between a user interaction (click, tap, key press) and the next frame update. The newest Core Web Vital. Good: ≤200ms | Needs work: 200–500ms | Poor: >500ms
The biggest cause in React/JavaScript-heavy sites: heavy JavaScript execution blocking the main thread. Every long task (>50ms) delays the browser's ability to respond to user input.
Fixes:
- -Break long JavaScript tasks into smaller chunks using setTimeout/requestIdleCallback
- -Move heavy computation off the main thread using Web Workers
- -Lazy load JavaScript bundles that aren't needed on page load
- -Audit third-party scripts (analytics, heatmaps, chat widgets) — these are often the worst culprits
The UK Context: Field Data vs Lab Data
Google uses field data (real user measurements via Chrome UX Report, or CrUX) for ranking purposes, not lab data from tools like Lighthouse. This is important because:
- -Lighthouse tests from a simulated mid-range device in a data centre
- -CrUX measures real UK users on real connections and real devices
- -A site can score 95+ on Lighthouse but fail field data CWVs if real users are on slower connections
You can check your CrUX data at pagespeed.web.dev. The field data section (if there are enough users) shows your real-world pass rates.
If your site doesn't have CrUX data yet (new site or low traffic), use a combination of Lighthouse and WebPageTest from a UK server (use WebPageTest.org with a London server).
The Conversion Impact: Real Data
Load time ↔ conversion rate:
- -0–1 second load: conversion baseline
- -1–2 seconds: -4.5% conversion
- -2–3 seconds: -9% conversion
- -3–5 seconds: -22% conversion
- -5–10 seconds: -38% conversion (Source: Google/Deloitte, 2022, still directionally accurate in 2026)
CLS impact:
- -Sites with good CLS (< 0.1) have 24% lower bounce rate than sites with poor CLS
- -High CLS causes form abandonment when buttons shift mid-interaction
Mobile-specific:
- -70% of UK web traffic is mobile
- -Mobile LCP is typically 40–60% worse than desktop
- -Google uses mobile performance for mobile ranking signals
How to Audit Your Site
Step 1: Run your homepage through pagespeed.web.dev. Note the three CWV scores and which ones are failing.
Step 2: Run the same URL through WebPageTest.org (London server, mobile device). Look at the waterfall and identify the largest blocking resources.
Step 3: Install the Chrome Web Vitals extension. Browse your own site and watch INP as you interact with it.
Step 4: Check Google Search Console → Core Web Vitals report. This shows real-user data grouped by URL type (homepage, article, product page, etc.).
The Priority Fix List for UK Websites
If you fail LCP: Image optimisation first, then server response time, then render-blocking resources.
If you fail CLS: Image dimensions second, font loading third, dynamic content fourth.
If you fail INP: Audit third-party scripts first, then JavaScript bundle size, then long tasks.
Tools to help:
- -Google Lighthouse (built into Chrome DevTools)
- -WebPageTest.org
- -Cloudflare (free CDN for UK businesses — dramatically improves LCP)
- -Squoosh.app (free image compression and format conversion)
- -web.dev/measure (Google's official CWV measurement tool)
When to Get Help
DIY CWV optimisation is feasible if you have developer access and the issues are primarily images and third-party scripts. If your issues are framework-level (poorly optimised React bundle, server-side rendering gaps, or complex CLS from a custom build), you need a developer who understands modern web performance.
Get a free website audit — we'll run a full Core Web Vitals analysis, identify your specific failing points, and estimate the conversion uplift from fixing them.
Related: Web Design Trends That Convert | Free Business Audit
