TL;DR
Core Web Vitals are Google's way of measuring whether your website actually feels fast and stable to a real person. They matter for two reasons: they're a ranking factor, and, more importantly, a slow or janky site quietly shoos visitors away before they ever enquire. One warning: that perfect score you got from a speed test might be quietly lying to you, because it measures a lab, not your actual customers. Here's what the numbers mean, and why they cost you money.
Strip away the jargon and Core Web Vitals measure three simple things about how your website feels to use:
Google bundles these into a pass-or-fail health check. It's not the mightiest ranking factor in the world, but it's a real one, and it's a tiebreaker: when two sites are otherwise neck and neck, the one that feels nicer to use wins.
Here's the part that matters more than rankings. Every extra second your site takes to load, you lose people. That's not a theory, it's just how humans behave: they're impatient, they're usually on a phone, and if your page hangs they hit back and try the next result, which is, of course, your competitor.
So a slow website doesn't merely rank a notch lower. It leaks leads right at the top of the funnel, before anyone reads your offer, sees your reviews, or fills in your form. You can have the sharpest copy and the best offer in your industry and still lose the sale because the page took four seconds to turn up.
That's why we treat speed as a conversion issue, not a nerd metric. It's the leakiest, most invisible hole in a lot of otherwise excellent websites.
A slow website doesn't lose you a ranking. It loses you the customer, before they ever see what you're selling.
Now the uncomfortable bit. You run your site through a free speed test, get a shiny 100 out of 100, and relax. Don't relax just yet.
Those instant scores are mostly a lab test: a simulated visit on a fast connection under ideal, laboratory conditions. Useful, but it isn't your actual customers. Google increasingly cares about field data, the real experience of real people on real phones and real networks, patchy mobile reception and all.
The two can disagree wildly. A site can score beautifully in the lab and still feel like wading through treacle to a customer standing in a car park on three bars of reception. If you only ever check the lab score, you might be congratulating yourself while your real visitors quietly bounce. Judge your speed by what your actual customers experience, not by a best-case simulation. Sorting out that gap is squarely technical SEO work.
Key takeaway
A perfect PageSpeed score measures a lab, not your customers. What matters is how fast your site feels to a real person on a real phone. Optimise for them, not for the test.
When a site is slow, it's almost always the same short list of offenders:
None of these are exotic. They're the accumulated cruft of a website that's been added to for years and never tidied. The fix is usually less "rebuild everything" and more "clear out the junk drawer".
You don't need to become a developer. You need the right things checked and fixed in the right order:
If your website is fundamentally old and creaky, sometimes the honest answer is that a faster foundation (a proper rebuild) pays for itself in recovered leads, which is where web design and speed meet. But for most sites, a focused clean-up gets you most of the way. Either way, speed isn't vanity. It's the difference between a visitor who stays and a lead you never knew you'd lost. It sits underneath everything else in SEO, because the fastest way to waste good rankings is to funnel people onto a slow page.
They're Google's measure of how fast and stable your website feels to a real person: how quickly the main content loads, how fast the page responds to taps, and whether things jump around while loading. They're a ranking factor and, more importantly, they affect whether visitors stay.
Yes to both. Speed is a genuine ranking factor and a tiebreaker, and more importantly a slow site drives visitors away before they enquire, so it quietly costs you leads regardless of rankings.
Not necessarily. Instant scores usually measure a lab simulation on a fast connection, not your real visitors. Google increasingly weighs field data from real people on real phones, so check that too before relaxing.
Compressing and right-sizing your images is usually the biggest win for the least effort, followed by removing unused plugins and scripts and sorting out hosting and caching.
About the author
Founder, Digital Marketing Adelaide
Matt Williamson is the founder of Digital Marketing Adelaide and Adelaide's face of AI Search. He leads the agency's AI search optimisation and training, and is a keynote speaker on how businesses get found in ChatGPT, Gemini and Google AI Overviews.
More about Matt