Improving Core Web Vitals – The PageSpeed Module
Your website loads sluggishly on the phone, and nobody can tell you why? That’s exactly what this module is for. It measures every page against a fixed standard – PageSpeed Insights 100/100/100/100 and green Core Web Vitals – and reports what misses the standard and why.
A measurement system with an uncomfortable standard
The module checks your pages regularly with PageSpeed Insights – mobile and desktop, all four categories: performance, accessibility, best practices, SEO. The target is 100 in every category. Not as a marketing promise, but as a test criterion for every single page and every component on it.
On top of that come the Core Web Vitals, Google’s metrics for real user experience: LCP measures the load time of the largest visible element, INP the response to input, CLS visual stability. INP replaced its predecessor FID in 2024. The thresholds for “good” are named in the Web Vitals documentation on web.dev: LCP up to 2.5 seconds, INP up to 200 milliseconds, CLS up to 0.1 – measured at the 75th percentile of real page loads.
Six causes that stay undetected without continuous measurement
A loading bar sells nothing. On the phone, in a dead spot between two subway stations, load time decides whether someone sees your offer at all.
Images too large
Delivered uncompressed and without modern formats, they delay the LCP the most.
Render-blocking JavaScript
Scripts that have to load before the first visible content delay the entire page.
Layout shifts (CLS)
Missing image dimensions and late-loading elements shift the content while the visitor is already reading.
Lab and field data confused
Lighthouse simulates a single load, CrUX shows real users. Reading only one source leads to wrong conclusions.
Third-party scripts
Consent tools and tracking pixels slow things down from outside, without appearing in your own theme.
Values tip over unnoticed
After every plugin update or new script – without anyone measuring it.
“I manage around 15 client projects in parallel, and PageSpeed follows the same pattern everywhere: optimised once, never looked at again. Two plugin updates later the values are red, and nobody noticed.”
— Viktor Pásztor, SEO freelancer in BerlinFive steps from measurement to approval
Read field and lab data
CrUX shows what real visitors experience. Lighthouse simulates a single load under fixed conditions. The module reads both sources, because they can diverge.
Check score and thresholds
Every page is checked against the target of 100/100/100/100 and the thresholds for LCP, INP and CLS – mobile and desktop.
Determine the cause
Images too large, render-blocking JavaScript, layout shifts – or a cause outside your own system: theme, consent and tracking scripts.
Propose a measure
Every measure lands as a proposal in your list, with the cause and the measured value.
Approval
Nothing is changed until you approve. Decisions about a theme change remain a separate project anyway.
DIY / pro / agency / automated
| Criterion | Do it yourself | SEO pro (manual) | Agency | How I do it |
|---|---|---|---|---|
| Costs | €0 – PageSpeed Insights is free | hourly rate, €100 with me | monthly retainer, depending on contract | €99–399 per month, depending on module scope, plus credit usage as consumed |
| Your time investment | high: measuring, interpreting, implementing | low, after briefing | low, after coordination | approvals, nothing else |
| Measurement rhythm | when you remember | occasional, per assignment | depending on contract | continuous, every page |
| Expertise needed | getting into metrics and causes | they bring it | they bring it | report in plain language, interpretation included |
| Comparability over time | screenshots instead of a measurement series | only with clean documentation | depends on the reporting | a continuous measurement series per page |
| Implementation | you yourself | per quote | via tickets | proposal, your approval, then implementation |
Four effects of continuous PageSpeed measurement
How much ranking a perfect score brings you depends on your website – I won’t invent percentages. What the module delivers:
A fixed standard
100/100/100/100 and green Core Web Vitals, mobile and desktop – not “it works, doesn’t it”, but a number
Fewer bounces
Fast pages keep mobile visitors in the purchase process to the end
Comparability
Every measurement lands in the series – you see what an update did to the values
Clear responsibilities
The report separates what’s down to your own system from what’s down to the theme or third parties
Questions about PageSpeed and Core Web Vitals
Does a score of 100 bring better rankings?
Not automatically. Load time is one ranking factor among many; content, structure and linking weigh more. That’s why I don’t give a ranking guarantee. What reliably improves are bounce rate and conversion on the phone.
Does the module work with my CMS?
Yes – with WordPress, Shopify, Shopware and TYPO3. Measurement happens from outside, including with PageSpeed Insights and GTmetrix. Only when implementing the measures does the respective system matter.
What if the cause lies in the theme or in third-party scripts?
Then that’s what the report says, with the measured value and the cause. Whether you change the theme or replace a consent tool, you decide – planned as its own project, not on the side.
My values are already green. Why measure continuously?
Because they don’t stay green. Every plugin update and every new script can tip the values. The measurement series shows the dip in the week it arises – not months later.
Back to the pillar
The PageSpeed module is a building block of MCP SEO, the ongoing SEO support delivered through modules. Two neighbours feed directly into it: the Redirects module clears up redirect chains that cost time on every load, and the Image SEO module fixes the most common PageSpeed cause of all – images that are too heavy.
The fundamentals without a module reference are under technical SEO, all prices clearly on the costs page.
Have your website’s PageSpeed checked
Send me your URL via the contact form. You’ll get a first assessment within 24 hours on weekdays – measured, not estimated.
→ Book your free consultation