Two and a half seconds is the number to hold on to. That is the target Google sets for Largest Contentful Paint, the point at which the main content of a page has appeared. Core Web Vitals are the three measures Google uses to score the experience: how fast the main content loads, how quickly the page responds to a tap, and how much the layout moves while it settles. For a B2B firm the cost of failing them is rarely a ranking penalty anybody notices. It is the enquiry form nobody reached.
The three Core Web Vitals numbers, and how a page passes
Each measure has a published target, set out in Google’s Web Vitals guidance.
| Measure | What it tracks | Target |
| Largest Contentful Paint | Time until the main content appears | Within 2.5 seconds |
| Interaction to Next Paint | Delay before the page responds to a tap or click | 200 milliseconds or less |
| Cumulative Layout Shift | How much the layout moves while loading | 0.1 or less |
Two details decide whether you pass. Google assesses at the 75th percentile of page loads, segmented across mobile and desktop, so a quarter of your slowest visits are set aside and the remaining three quarters have to hit the target. And a page passes only where all three measures meet their targets, so one failing measure fails the page.
The figures come from real visits on real devices and connections, which is why a site that feels quick on the office fibre connection can score badly on a phone in a car park.
Speed shows up in leads before it shows up in rankings
There is more published evidence here than most agency pitches bother to cite, and it is worth being specific about what it does and does not prove.
Google’s own performance guidance reports that the BBC ‘lost an additional 10% of users for every additional second their site took to load’, and that Redbus improved its Interaction to Next Paint and increased sales by 7%. Google’s collected Core Web Vitals case studies add Vodafone, where a 31% improvement in LCP produced 8% more sales, and Agrofy Market, an agricultural marketplace selling business to business, where 70% better LCP correlated to a 76% reduction in load abandonment.
Every one of those is a high-traffic consumer or marketplace site, and none is a forty-person UK engineering firm. The percentages do not transfer, and no credible benchmark exists at that scale. What transfers is the mechanism: a slow page loses the visitor before they act, and the load-abandonment figure is the one that names it directly.
That mechanism bites harder at low traffic, not softer. A site receiving a few hundred visits a month has fewer chances to lose, and a lost visit on a service page is a conversation that never starts. It is why page speed sits inside our web design and maintenance work instead of being sold separately, and why it matters most on paid traffic, where the benefits of PPC for businesses rest on the page the click lands on.
Where the seconds go on a B2B site
Four causes account for most poor scores we see on small business websites, and none of them requires a rebuild.
Images that are too large. A hero photograph exported at 4000 pixels wide and served to a phone is the commonest single problem. Correctly sized and compressed, the same image is a fraction of the weight.
Too many plugins. WordPress sites accumulate them, and each can load its own stylesheet and script on every page whether that page uses it or not. Sites carrying dozens of active plugins are common in our experience, and a good proportion of them do nothing.
Third-party scripts. Chat widgets, analytics, tag managers, embedded maps and tracking pixels all load code from elsewhere, and your page waits. This category grows quietly, because each addition is somebody’s reasonable request.
Fonts and layout shift. A custom font loading late redraws every line of text, and an advert or cookie banner inserted above existing content pushes it down as the visitor reaches for a link. Both hurt the layout score and both are ordinary maintenance to fix.
Testing your own site takes about ten minutes
Put your URL through Google’s PageSpeed Insights and read the field data at the top of the report. Google’s PageSpeed Insights documentation states that it reports real users’ experiences ‘over the previous 28-day collection period’, and reports the 75th percentile for all metrics. The laboratory score underneath is a simulated load on a single device, so it moves around; the field data is the historical record of what your visitors met.
Test three pages. Your home page is usually the fastest, because it gets the most attention, so test a service page and a blog post as well. Those are where search traffic lands, and they tend to carry the heaviest images and the least attention.
Then compare mobile against desktop. Where mobile fails and desktop passes, image weight or a script blocking the page accounts for most of what we see, and both are addressable without touching the design.
Fixing this is maintenance, not a redesign
The instinct when a site scores badly is to replace it. Occasionally that is right, where the site is genuinely old or built on something unmaintainable. Mostly it is an expensive answer to a maintenance problem.
Four steps resolve a large share of failing scores:
- Compress and correctly size the images.
- Remove the plugins nothing uses.
- Audit the third-party scripts and drop the ones no longer earning their place.
- Set explicit dimensions on images so the layout stops moving.
That keeps the content and the search history you already have. The exception is a site whose theme loads its own heavy framework on every page: where the foundation is the problem, patching around it costs more than replacing it once, and an honest assessment should tell you which situation you are in.
Where to start
Run the test on three pages, note the mobile field data, and look at your image weights first. That sizes the problem and tells you whether this is an afternoon of work or a project.
Every website we build includes hosting, ongoing maintenance and basic search optimisation as standard. If your site is losing enquiries and you want to know which of the four causes is responsible, ask us for a website redesign and SEO health check, which is where we look at an existing site and tell you whether the fix is maintenance or a rebuild. Our digital marketing articles cover the wider picture.