Your PageSpeed Score Is Not Your Speed
The number in Google PageSpeed Insights is a laboratory test: one simulated mid-range phone on a throttled connection, run once. It is a useful diagnostic and a terrible target.
What Google actually uses for ranking is field data — the loading experience of real visitors on real devices over the past month, which appears at the top of the same report when your site has enough traffic. A site can score 45 in the lab and pass on field data, or score 90 and still feel slow to someone on patchy mobile data in the evening.
So the honest question is not “what is my score” but “what do my visitors experience”. The two usually move together, which is why the score is worth fixing — but chasing the last ten points is where money stops working.
What Actually Moves the Number, In Order
- Images. Almost always the largest thing on the page and almost always uploaded at the wrong size. Our guide on image optimisation for a website covers this on its own.
- Caching. The difference between building the page for every visitor and serving one that is already built.
- Plugin and script load. Every plugin adds weight, and the ones added for a campaign three years ago are still loading on every page.
- Fonts and third-party scripts. Chat widgets, analytics, ad pixels — each one is a request to somebody else’s server before your page finishes.
Hosting sits underneath all of it and is sometimes genuinely the limit, but often is not — we check before recommending a move, and our hosting and domain support page covers that side if it turns out to be the cause.
When Not to Bother
If your site already loads in about two seconds on a phone and its field data is passing, further optimisation is a poor use of money whatever the lab score says. Spend it on what the site says instead. The exception is if you are running Google Ads, where landing page experience feeds Quality Score and a faster page can genuinely lower what you pay per click.
If you would rather try it yourself first, our guide on how to speed up a WordPress website covers the same steps in order.
For a plain-language map of everything that affects load time, from Core Web Vitals to caching and hosting, read our website speed guide.
What a Page Speed Optimisation Service Actually Does
A page speed optimisation service is not one setting or one plugin. It is a short investigation followed by a list of fixes in the order that helps most. On a typical WordPress site, the work looks like this:
- Measure first. Record the current field data and lab results for the pages that matter — usually the home page, the main service or product pages and the contact page — so every change can be compared against a starting point.
- Find the heaviest things on each page. Large images, unused scripts, fonts, sliders and embedded videos are listed by how much they cost.
- Fix images. Resize them to the size they are actually shown, convert to WebP, and lazy-load anything below the first screen.
- Set up caching properly. Page caching, browser caching and, where the hosting supports it, object caching. Settings depend on the server — LiteSpeed Cache and other plugins need different choices.
- Reduce plugin and script load. Remove what is not used, and stop scripts loading on pages that do not need them.
- Check the hosting. If the server itself is slow to respond, no front-end fix can hide it. Does hosting affect website speed? explains how to tell.
- Re-measure and report. The same pages are tested again, and you get the before and after numbers along with what was changed.
Core Web Vitals in Plain Language
Google measures page experience through three Core Web Vitals. Knowing what each one means makes a speed report much easier to read:
| Metric | What it measures | Common cause when it fails |
|---|---|---|
| LCP — Largest Contentful Paint | How quickly the main content (often a hero image or headline) appears | A large hero image, slow server response, render-blocking CSS |
| INP — Interaction to Next Paint | How quickly the page reacts when someone taps or clicks | Heavy JavaScript from sliders, page builders, chat and tracking scripts |
| CLS — Cumulative Layout Shift | How much the page jumps around while loading | Images without dimensions, late-loading banners, web fonts swapping in |
Our Core Web Vitals guide goes deeper, and if Search Console says your site failed the assessment, start with Core Web Vitals assessment failed.
Speed on WooCommerce Stores
Stores are slower than brochure sites for structural reasons: product pages carry many images, the cart updates itself in the background, and checkout pages cannot be cached the way normal pages can. That means store speed work is different. Product images, cart fragments and the database matter more, and caching has to exclude the cart and checkout. If your shop is the slow part, see why a WooCommerce site is slow and slow WooCommerce checkout. Complex WooCommerce and multi-site setups are quoted separately after a review.
Speed and Google Ads Landing Pages
If you pay for clicks, a slow landing page costs you twice: visitors leave before the page loads, and landing page experience is one of the things Google Ads looks at. A landing page usually needs a narrower kind of speed work — one page, one form, as little script as possible. Read landing page speed for Google Ads, or see our Google Ads landing page design service if the page itself needs rebuilding.
Before You Hire Anyone for Speed
A few minutes of your own testing makes any quote easier to judge. Run your main pages through our free website speed test, then read the result with how to read a PageSpeed Insights report. When you compare providers, ask what they will measure, which pages they will test and what they will change — our article on hiring someone to speed up a website lists the questions worth asking, and website speed optimisation cost explains what drives the price.