Slow website: how speed costs you sales and what to fix first (Core Web Vitals in plain English)
Why a slow website costs you leads and sales. Core Web Vitals in plain English: LCP, INP and CLS, how to measure them yourself and what to speed up first.
Short answer: a slow website loses visitors before they even see your offer, and with them go leads and sales. Three things usually eat the speed: heavy images, a slow server response with no caching, and third-party scripts (chat widgets, pixels, embeds). Start by measuring in PageSpeed Insights, then fix whatever slows the first screen on mobile.
Owners usually check their site on an office computer with good internet, and it seems fine. A customer opens it on a phone on the bus after tapping an ad and stares at a blank screen for several seconds. Some close the tab and go to a competitor, and you have already paid for that click. Here is how to judge your site speed without technical knowledge and what to fix first.
Core Web Vitals in plain English
Core Web Vitals are three Google metrics that describe how a site feels to a person. Google measures them from real Chrome users and uses the 75th percentile, so a metric must be good for at least three visits out of four. The thresholds are published on web.dev.
| Metric | What it measures | Good | Typical causes of a poor result |
|---|---|---|---|
| LCP, Largest Contentful Paint | When the main element of the first screen appears: hero image, photo, headline | 2.5 s or less | Heavy image or slider, slow hosting, no caching, render-blocking CSS and JS |
| INP, Interaction to Next Paint | How quickly the site responds to a tap: button, menu, filter, “add to cart” | 200 ms or less | Too much JavaScript, third-party scripts, heavy page builders |
| CLS, Cumulative Layout Shift | Whether the page jumps while loading so people tap the wrong thing | 0.1 or less | Images without dimensions, banners and cookie bars injected at the top, font swaps |
The difference between field and lab data matters. Field data comes from real visitors on their own phones and connections, and that is what Google uses. Lab data is a single test under fixed conditions; it is useful for finding causes and checking a fix quickly, but it does not show how people actually experience the site.
INP replaced the older FID metric in March 2024. If a vendor still reports FID, the report is out of date.
How speed costs you sales
- The visitor gave up. You paid for the click and they left before seeing the offer.
- The visitor waited but got annoyed. The catalogue filter takes seconds, “Buy” responds late, and the cart is abandoned. Other reasons shoppers leave are in abandoned carts: what actually works.
- The page jumps. They tap “Call” and a different element slides under their finger.
- Ads cost more. Google counts landing page experience in Quality Score, and a slow page converts worse. Both push up cost per lead; see why ads cost more but bring fewer leads.
Check it yourself in 10 minutes
- Open PageSpeed Insights and test your home page, a service or product page, and the cart.
- Read the top block first, “Discover what your real users are experiencing”. That is field data from Chrome over the last 28 days. If it says there is not enough data, the site has few Chrome visitors; use the lab test below it as a guide.
- Switch to Mobile. Most small-business traffic comes from phones.
- In the diagnostics, find which element is the LCP and check its size. A multi-megabyte photo on the first screen is the most common finding.
- In Google Search Console, open the Core Web Vitals report to see which groups of pages are slow. If the site is not verified there yet, or hardly any pages are indexed, deal with that first: see website not showing up on Google.
- Open the site on your phone over mobile data in a private tab and count the seconds until you can read and tap.
What to fix first: the order that pays off most
1. Images
The most common cause and the cheapest to fix. A 3–5 MB photo from a phone or stock library used as a hero banner already makes a slow site. Use WebP or AVIF, size images to their real display width, compress them, and lazy-load anything below the first screen. Do not lazy-load the hero image; give it loading priority instead.
2. Caching and server response
If the server takes a long time before sending the page, other optimisations help little. For WordPress, use page caching via a plugin or the host, a current PHP version, and proper hosting instead of the cheapest shared plan. A CDN helps if your visitors are spread across countries.
3. Third-party scripts
Live chat, several ad pixels, a reviews widget, a map, embedded YouTube videos, a discount pop-up: each script adds weight and takes browser time. Check which ones you actually use. Remove the rest or delay them until the visitor interacts. Duplicate analytics tags are a separate problem we often find in audits.
4. The first screen
A five-slide carousel or a background video on the first screen is expensive for speed and rarely pays for itself in sales. One good image and a clear headline with a button often work better.
5. Plugins, builders and apps
Each WordPress plugin or Shopify app can load its own styles and scripts on every page, even where it is not needed. Auditing plugins and apps often gains more than any “speed plugin”.
6. Layout stability (CLS)
Set dimensions on images and videos, reserve space for banners and the cookie bar, and load fonts so text does not jump when they arrive.
Platform specifics
- WordPress and WooCommerce: hosting, page caching, the number of plugins and heavy page builders matter most. In WooCommerce, check filtered catalogue pages and the cart separately: they are not cached like ordinary pages and slow down first on large catalogues.
- Shopify: the platform handles servers and the CDN, so the main gains are in the theme and apps. An app can keep loading scripts even after you uninstall it if its code stays in the theme.
- Webflow and Squarespace: hosting is also on the platform side. Speed is usually eaten by large images and videos, animations and embedded third-party widgets.
- React or Next.js sites: make sure the first screen is rendered on the server and the JavaScript the browser must run before the first interaction is kept to a minimum.
What not to do
- Do not install three optimisation plugins at once. They conflict, and the site can break or get slower.
- Do not chase a lab score of 100. The goal is good Core Web Vitals for real visitors and a site that is pleasant to use, not a number on a screenshot.
- Do not change everything on the live site at once. Work in steps on a staging copy and measure after each one.
- Do not expect field data to change overnight. The lab test shows improvements immediately; real-user data rolls over 28 days.
From our own work
The building materials store on WooCommerce carries thousands of products, and a catalogue that size slows down without deliberate speed work. We set up caching and speed optimisation (WP Rocket) and technical SEO, and the catalogue from the accounting system reaches the site without manual entry.
For images on client WordPress sites we use our own plugin, Complete Image Optimizer. It converts new and existing images to WebP right on the server, with no third-party subscription. It is one of the first steps we take in speed work and ongoing support.
When to bring in a specialist
- Core Web Vitals in Search Console are “Poor” or “Needs improvement” and you cannot tell what is slowing the site;
- the site is still slow after caching and image plugins;
- the site runs on a heavy page builder or an old theme, and every change breaks something else;
- a store with a large catalogue lags on filters, search or the cart;
- you are about to run ads and do not want the budget to burn on a slow page.
Symptoms and causes are covered on the slow website page. Start with a technical audit (from €300): Lighthouse on key pages, a review of the theme, plugins and hosting, and a prioritised report with an estimate. To keep the site in shape over time, there is WordPress support (from €100 per month). If the foundation itself is outdated, a redesign on a fast foundation may be the better route; keep our redesign SEO checklist at hand so the move does not cost you rankings. Send us your URL and we will tell you what slows it down the most.
FAQ
What PageSpeed score is good enough?
Focus on Core Web Vitals from real-user data rather than the score: LCP under 2.5 s, INP under 200 ms, CLS under 0.1. The lab score varies between runs and depends on test conditions.
PageSpeed says my site is slow, but it loads fast for me. Why?
You open it on a powerful computer with fast internet and often a warm cache. PageSpeed simulates a mid-range phone on a slower connection, and the field data reflects your real visitors.
Will a caching plugin fix it?
For WordPress, page caching is one of the first steps and often cuts server response time noticeably. It will not fix heavy images, a hero slider or a dozen third-party scripts.
Does speed affect Google rankings?
Core Web Vitals are part of the page experience signals Google uses, but relevance matters more. In practice, speed has a bigger effect on whether visitors wait for the page and enquire.
How much does it cost to speed up a website?
A technical audit with the causes and an estimate starts at €300. Images, caching and removing unneeded scripts often deliver most of the gain. If the problem is an outdated theme or page builder, a redesign from €1,200 can be the better investment.