There's a silent metric that keeps deciding who ranks and who doesn't, and who converts and who doesn't. It doesn't get as much airtime as GEO or AI Overviews, but in 2026 it carries exactly the same weight it did five years ago. If anything, more. And most websites still fail it.
We're talking about Core Web Vitals: the three performance metrics Google measures for every site and which, alongside content relevance, determine where you appear. LCP, INP and CLS — three acronyms you've probably heard before, and worth revisiting because what's happening beneath them is moving fast.
Why performance still decides rankings (even when Google talks about it less)
Over the past two years, Google's public narrative has focused on AI, AI Overviews, generative search. That makes sense — it's the new thing. But in the background, Core Web Vitals have continued to be weighted exactly as before: a site that takes three seconds to render its main content loses positions to one that loads in a second and a half, everything else equal.
The difference from five years ago is that the bar has raised. An LCP of 2.5 seconds, which was once "reasonable," now sits on the yellow line. The average among high-performing sites has moved to 1.8s. Sites still sitting at 3s are starting to lose visibility without understanding why.
The conversion impact: more important than the ranking
There's a number that interests us more than the ranking: every extra second of load time on mobile drops conversion rates between 7 and 20% depending on the sector. We see it with clients constantly. A site with a 3s LCP on mobile converts 30–40% worse than its equivalent optimized to 1.5s. Not because it looks bad or has poor content — because users get impatient and leave before they reach the form.
This is critical for SMBs and local businesses. 70% of traffic is mobile, 70% of mobile traffic is on average connections, and 70% of purchase decisions are abandoned when there's friction. The cheapest friction to detect and the easiest to fix is performance.
Why most WordPress sites fail the test
This isn't an opinion — it's what the data shows. The typical stack: a WordPress with a heavy premium theme, six essential plugins, a visual page builder, Google Fonts loading three families, an embedded external chat widget, an analytics pixel loading third-party scripts. That combination produces sites that, no matter how much you optimize, start with a structural handicap.
It can be fixed, yes. With caching plugins, lazy loading, a CDN, unused CSS purging. But there's a point where optimizing the Frankenstein costs more than building the site on the right technology from scratch.
When to rebuild and when to just optimize
This is the question every client asks. Our honest answer: it depends on three things.
If your current site is less than three years old and was built with some discipline — a lightweight theme, minimal plugins — it can be optimized to a reasonable PageSpeed score (85+) without rebuilding. We do it often. It's precise surgical work, and it requires the right hosting.
If your site is more than five years old, has accumulated plugins from multiple previous agencies, and runs on shared hosting that takes 3 seconds just to deliver the HTML — optimizing is patching a car that needs a new engine. In that case, a migration to a modern stack (Next.js, Astro, or headless WordPress done well) is more efficient and longer-lasting.
If you're launching a new site — there's no question. Building it with performance as a first-class concern from day one saves years of technical debt. And it doesn't cost meaningfully more.
An example of what's achievable
The Menorca Studio website scores 98 out of 100 on PageSpeed for mobile. LCP under one second. Minimal INP. Zero CLS. That's not a feat — it's a site built with the right technology from the start and with images consciously optimized. The result isn't just that Google positions it well: it's that a client opening it on their phone while comparing architecture practices sees the portfolio loaded before the doubt to close the tab ever forms.
The right frame
Core Web Vitals aren't "technical SEO." They're user experience expressed in numbers. Businesses that take care of them convert more and rank higher, without having changed a single word of content. Those that ignore them pay double in ads to compensate for the silent hole.
In 2026, with more competition for attention than ever, performance isn't optional. It's the foundation. And fortunately, it's one of the things that can be fixed properly and fixed quickly — as long as it's done with discipline.
At A-Digital we've spent years building sites where performance is a design consideration from the first commit, and optimizing existing ones. Clients in Menorca, mainland Spain, and Europe. The pattern is always the same: those who get this right sell more. There's no magic — there's a very clear curve between milliseconds and revenue.
We'll run a free performance diagnosis of your site: which metrics you're failing, why, and whether you need to optimize or migrate. Concrete numbers, not gut feelings.
Request your free audit →