TL;DR
Both a ranking factor and a conversion problem, but not equally, and the honest ordering is the opposite of how it gets sold. Forty-eight per cent of mobile origins pass all three, so half the web fails and ranks fine. It is a real signal and a weak one. The conversion case is the stronger one.
Short answer: both, but not equally, and the honest ordering is the opposite of how it usually gets sold.
It is a real ranking signal and a weak one. It is a strong conversion factor. Most of the pitches I see reverse that, because “Google will rank you higher” closes better than “fewer people will leave.”
The number that settles the ranking question
HTTP Archive’s 2025 Web Almanac found that as of June 2025, 48% of mobile origins pass all three Core Web Vitals: 62% good on LCP, 81% on CLS, 77% on INP. On desktop it is 56%.
So more than half the mobile web fails, and a great deal of it ranks perfectly well, including plenty of sites sitting above you right now. That is not an argument for ignoring it. It is an argument against believing it is the reason you are invisible.
If a site is not ranking and its vitals are poor, the vitals are rarely the cause. They are usually one symptom of a site nobody has maintained, which is the actual finding.
Where it does decide things
Two situations, and they are narrower than the pitch suggests.
As a tiebreaker. Between two pages Google considers comparably relevant, page experience can be the thing that separates them. That is a real effect and it is small. It matters most in competitive commercial results where the top few are genuinely close.
When it is catastrophic rather than mediocre. There is a difference between an LCP of 3.1 seconds and one of 11 seconds. The first is a number on a report. The second is a page most people never see, because they left.
The conversion case, which is stronger
This is the argument I actually make to clients, and it does not need Google in it at all.
A person who leaves before your page renders did not evaluate your offer. They are not a lost ranking, they are a lost sale, and you paid for them to arrive whether the traffic was organic or not. On an ecommerce store the arithmetic is direct enough that you do not need a study. Put the number of sessions next to the bounce rate on your slowest template and it makes itself.
CLS is the one I would push hardest on and it is usually treated as the least important. A layout that shifts under someone’s thumb as they tap is not a metric problem, it is a person tapping the wrong button on a checkout.
What the platform data tells you about effort
The Almanac’s CMS chapter puts mobile Core Web Vitals pass rates at 45% for WordPress, 74% for Wix and 85% for Duda.
WordPress is not slower by nature. It is slower by accumulation. It is the platform where you can install anything, so people do, and roughly 60% of WordPress sites are running a page builder on top. The pass rate is what years of additive changes add up to.
Which tells you where the effort goes. On a managed platform, performance work is mostly image handling and third-party scripts. On WordPress it is those plus a real audit of what has accumulated, and that second part is the bigger number.
What I would do, in order
- Use field data, not lab scores. Search Console’s Core Web Vitals report and the Chrome UX Report describe real visitors. Lighthouse describes a simulated device in a data centre and will happily give you a score that has nothing to do with your customers.
- Fix by template, not by page. One product template covers every product.
- Start with images and third-party scripts. Between them they are most of the problem on most sites, and neither needs a developer.
- Then stop. Getting from poor to good is worth doing. Getting from good to excellent is a hobby, and there is almost always something on the site costing more.
The short version
- 48% of mobile origins pass all three. Half the web fails and ranks fine.
- It is a real ranking signal and a weak one: a tiebreaker between comparable pages.
- Catastrophic is different from mediocre. Eleven seconds is a lost visitor, not a metric.
- The conversion argument is stronger and does not need Google in it.
- CLS is under-rated. A shifting layout is someone tapping the wrong button.
- WordPress passes 45%, Duda 85%. That gap is accumulation, not the platform.
- Field data, fix by template, images and scripts first, then stop.
If a site should be ranking and it isn’t, that’s the work I do. Technical SEO covers rendering, performance and the layer underneath them. If you’re not sure which of several plausible problems is costing you, that’s what an SEO audit is for.
