Most small business owners we speak to assume their website is fast enough. It loads on their office broadband, it looks fine on their phone, and nobody has ever complained. Core Web Vitals for small business sites tell a rather different story, because Google measures real visitors on real connections, not the version you see after your own browser has cached everything.
Speed has been a ranking factor for years, but 2026 changed how much it counts. Google's March core update pulled the three Core Web Vitals into a single performance signal that now sits alongside content quality rather than quietly behind it. For competitive local searches, that shift is often enough to separate two otherwise very similar businesses.
The gap is wider than most people expect. Roughly a third of websites currently pass all three metrics, which means two thirds are leaving something on the table. If your competitors sit in that failing majority, fixing your own site is one of the few technical wins still genuinely available to a small business on a modest budget.
Largest Contentful Paint, or LCP, measures how long the main content of a page takes to appear. Google wants this under 2.5 seconds. On most small business sites the largest element is the hero image or a big headline block, so a heavy uncompressed photo at the top of your homepage is usually the first thing worth fixing.
Interaction to Next Paint, or INP, measures how quickly your site responds when someone taps a button, opens a menu or starts filling in a form. The target is 200 milliseconds. This is now the most commonly failed metric across the web, and it is almost always caused by too much JavaScript running quietly in the background.
Cumulative Layout Shift, or CLS, measures how much the page jumps around while it loads, and should stay below 0.1. Google grades all three at the 75th percentile of real visits, so you pass only when the large majority of visitors get a good experience. Our web design team in Wales builds to these thresholds from the first wireframe rather than retrofitting them later.
Rankings are only half the story. Sites that pass all three thresholds typically report conversion improvements in the region of 15 to 30 percent, and pulling load time down from three seconds to one second has been shown to lift conversion rates by as much as 27 percent. For a business taking enquiries through a form, that is real money.
Paid traffic suffers too. Every visitor who bounces before your page renders is a click you have already paid for, and Google factors landing page experience into ad quality. A slow site quietly raises your cost per lead across every channel at once, which is why performance problems often show up first as a rising ad budget rather than a falling ranking.
Then there is the compounding effect on search. Slow pages get crawled less efficiently, earn fewer links and lose the engagement signals that support organic growth, so the damage builds steadily over months. We treat performance as a foundation of any SEO campaign in Wales rather than a separate technical job bolted on at the end.
Start with Google PageSpeed Insights and pay attention to the top section, which shows field data from real Chrome users over the previous 28 days. The lab score underneath is a simulation, useful for diagnosis, but only the field data reflects what Google actually uses. If a page has too little traffic to show field data at all, that is worth knowing too.
Next, open the Core Web Vitals report in Google Search Console. It groups your URLs by issue and by template, so you quickly see whether the problem is your homepage, your service pages or every blog post on the site. Fixing one template usually fixes dozens of pages at once, which makes this the fastest route to a meaningful improvement.
Always test on mobile first, because that is where most small business traffic arrives and where the thresholds are hardest to hit. Check your homepage, your main service or product page and your contact page as a minimum, then repeat the whole exercise monthly. Plugin updates and new tracking scripts undo good work far more often than owners realise.
Server response time is the foundation, and cheap shared hosting is where a lot of small business sites lose their first full second. Moving to better hosting, running a current version of PHP and enabling proper server-side caching often produces a bigger improvement than weeks of front-end tinkering. It is the least glamorous fix and usually the most effective one.
After that, look hard at your hero area. Serve images in modern formats such as WebP, size them for the space they actually occupy rather than uploading straight from a camera, and preload the single image that appears first. Sliders and background videos above the fold remain the two most common causes of a failing LCP score.
Product pages deserve extra attention, because every image, review widget and related-items carousel adds weight to the page a customer is closest to buying from. When we work on an e-commerce site, template performance work almost always pays for itself faster than any other single change we could make.
INP problems come from JavaScript, so start by auditing exactly what is running. Chat widgets, heat mapping tools, duplicate analytics tags, review carousels and abandoned marketing pixels all compete for the same main thread. Most sites we audit are carrying at least three scripts nobody can justify any more, and removing those costs nothing at all.
CLS is usually simpler to solve. Give every image and embed explicit width and height attributes, reserve space for banners and cookie notices instead of letting them push content down the page, and load custom fonts in a way that avoids a visible swap. These are small changes a developer can generally make in an afternoon.
One warning is worth repeating. A caching plugin alone will not save you, because caching helps LCP but cannot undo a bloated page builder, a theme loading twelve stylesheets or third-party scripts blocking interaction. If your site still fails after the obvious fixes, the honest answer is often that the underlying build needs replacing.
Our advice is to sequence the work rather than attempt everything at once. Fix hosting and images first, audit and remove unnecessary scripts second, and only then consider a rebuild. Most small business sites can move from failing to passing on those first two steps alone, and you will see the difference in Search Console within a few weeks.
Decide early whether you are repairing or replacing. If your site is more than four years old, built on a heavily customised theme and carrying plugins nobody remembers installing, a fresh build to modern standards is usually cheaper than an endless sequence of patches. A website that is fast by design tends to stay fast.
At Primedia we build sites that pass Core Web Vitals for small business budgets, and we audit existing sites honestly before recommending anything at all. If you are not sure whether speed is quietly costing you enquiries, ask our team for a performance review and we will tell you plainly what is worth fixing and what is not.
A page passes when Largest Contentful Paint is 2.5 seconds or less, Interaction to Next Paint is 200 milliseconds or less, and Cumulative Layout Shift stays below 0.1. Google grades these at the 75th percentile of real visits over a rolling 28-day window, so the large majority of your visitors need a good experience, not just a lucky few.
It helps, but it is not a magic switch. Performance acts as a tie breaker between pages of similar quality and relevance, so a fast page with weak content still will not rank well. The most reliable gains come from fixing performance on pages that already have solid content and genuine search demand behind them.
Basic work such as compressing images, enabling caching and removing unused scripts is often a few hundred pounds plus VAT and can be completed in days. Deeper problems, including a bloated theme or a heavy page builder, may need a rebuild. An audit first tells you which situation you are actually in before you commit any budget.
A caching plugin usually improves loading speed and can move LCP into a passing range, but it cannot fix layout shift or slow interactions caused by third-party JavaScript. Treat it as one useful tool rather than a complete solution, and always retest with real field data a few weeks after installing one.
Monthly is sensible for most small businesses, with an extra check after any significant change such as a new plugin, a redesigned page or a fresh tracking script. Performance drifts quietly, and problems are far cheaper to fix in the weeks after they appear than after months of lost traffic and enquiries.
Had your Google Business Profile suspended? We explain why it happens in 2026, how to fix the cause, and how to appeal so your listing comes back fast.
Core Web Vitals for small business explained: what LCP, INP and CLS mean, why they affect rankings and sales, and how to make your site faster in 2026.
Not sure which social media platforms for small businesses actually pay off in 2026? We share how to pick one or two channels and make them work.
