Website speed: how to test it and make your site faster (Core Web Vitals)

1. Why website speed affects sales and Google rankings

Most visits to business websites now come from phones, often on mobile data, on the move, between one thing and the next. If the page shows a white screen for several seconds, some people go back to the results and click a competitor. There is no error message — the enquiry simply never arrives.

Google measures that same experience on real Chrome users and counts it towards its assessment of the page as Core Web Vitals. Speed will not beat better content, but between two similar pages it can decide which ranks higher.

2. How to test website speed — step by step

Two free Google tools are all you need:

1. 1.Go to pagespeed.web.dev, paste the page address and run the analysis. Start with the Mobile tab — it matters more.
2. 2.Look first at the top section with real-user data from the last 28 days. If the site has too little traffic it will not be there — then the lab test below is what you have.
3. 3.Check the lab score (0–100) and the Diagnostics list: specific things to fix, sorted by how much time they cost.
4. 4.Repeat the test for the two or three most important pages — services, pricing, contact — not just the homepage.
5. 5.In Google Search Console, open the Core Web Vitals report. It shows which groups of pages have a problem, based on data from the whole site.

Real-user data and the lab test often disagree. The lab simulates an average phone on a slow connection and is for finding causes; field data shows what actually happens — and that is what Google looks at.

3. Core Web Vitals in plain English: LCP, INP and CLS

The three metrics Google uses answer three simple questions:

• —LCP (Largest Contentful Paint) — when does the main content appear?: The time until the largest element on screen is shown, usually the header image or the main heading. Good: up to 2.5 seconds.
• —INP (Interaction to Next Paint) — does the page respond straight away?: How long the page hesitates after a button tap, opening the menu or typing. Good: up to 200 milliseconds.
• —CLS (Cumulative Layout Shift) — does the content jump around?: How much elements move while loading — that moment when you go to tap a button and a banner slides into its place. Good: up to 0.1.

To pass, each of the three has to be in the green range for at least 75% of visits. An average is not enough — what counts is the experience of the slower share of your visitors.

4. What usually slows a business website down

On small and mid-sized business sites the causes repeat surprisingly often:

• —Heavy images: a photo straight from the camera is several megabytes and 6,000 pixels wide, shown on a screen 400 pixels wide. The single most common cause of poor LCP.
• —Third-party scripts: chat, ad pixels, maps, a reviews widget, an embedded video. Each one is small, but together they can freeze a phone for seconds and ruin INP.
• —Too many plugins and a heavy theme: in off-the-shelf systems every plugin adds its own files to every page — including pages where it does nothing.
• —A header slider: several large images rotating on load means several times more data to download before anything is visible.
• —Fonts and elements with no reserved space: fonts loaded at the last moment and images without set dimensions push the content around and damage CLS.
• —A slow server: the cheapest shared hosting with no caching can think for a second before sending the first byte.

5. How to speed up a website: what you can do yourself and what needs a rebuild

Fixes for a single afternoon

• —Convert images to WebP or AVIF and resize them to the size they are actually displayed at.
• —Turn on lazy loading for images below the first screen — but not for the header image, because that one decides LCP.
• —Remove plugins and scripts nobody uses, and load the chat after a delay or on the user's first interaction.
• —Replace the header slider with one good image.
• —Enable caching and compression on the server if your hosting allows it.

When fixes are not enough

If the site still takes several seconds on a phone after those steps, the problem usually sits deeper: in a theme that loads everything everywhere, in dozens of interdependent plugins, or in a structure that cannot be slimmed down without a rewrite. At that point further patches cost more than a light site built from scratch — prerendered, with images prepared at build time and scripts loaded only when they are needed.

6. Common questions about website speed

What is a good PageSpeed Insights score?

90–100 is green, 50–89 orange, below 50 red. Field data matters more than the number, though: if LCP, INP and CLS are green, the page passes Google's assessment even with a lab score of 85.

Why is the mobile score so much lower than desktop?

The mobile test deliberately simulates an average phone with a slower processor on a weaker connection. It is not a flaw in the tool — that is how the site looks to a large share of your customers.

Is a caching plugin enough to speed up a website?

It helps with a slow server, but it will not shrink images or remove heavy scripts. With poor LCP and INP, caching alone is rarely enough.

7. Test your own site

Run PageSpeed Insights on your homepage and your most important service page, on the Mobile tab. If any of the three metrics is red, you know where to start.

The sites I build treat speed as a starting assumption, not a fix at the end. Send me your site's address and within 48 hours you'll get a free quote, along with what slows it down most and whether fixes will do or a rebuild makes more sense.