HomeMCP SEO → PageSpeed
Module

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.

What it is

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.

Load timeline: FCP, LCP and TTI Schematic example – not a real measurement 0 s 0,5 s 1 s 1,5 s 2 s 2,5 s 3 s FCP First Contentful Paint 0,8 s LCP Largest Contentful Paint 1,9 s TTI Time to Interactive 2,6 s viktorpasztor.de · MCP SEO · PageSpeed module
Schematic example of a load timeline with the milestones First Contentful Paint (FCP), Largest Contentful Paint (LCP) and Time to Interactive (TTI) – not a real measurement of this or any specific website.
Why it’s a problem

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 Berlin
Measurement: one-off vs. continuous Why a score without repetition says nothing about today. One-off measurement Start: measured ? No further measurement point for months Continuous measurement (module) Update detected reported fixed Every page in the measurement series – deviations get noticed in the week they arise. viktorpasztor.de · MCP SEO · PageSpeed module
Schematic illustration: without continuous measurement, a deterioration stays unnoticed. The module detects the drop in the week it arises and reports the cause.
How to solve it

Five steps from measurement to approval

The process in five steps 1 2 3 4 5 Read field+ lab data Checkscore Determinecause Proposemeasure Approval
Only after your approval is a measure implemented – causes in the theme or with third parties remain your own decision.
1

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.

2

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.

3

Determine the cause

Images too large, render-blocking JavaScript, layout shifts – or a cause outside your own system: theme, consent and tracking scripts.

4

Propose a measure

Every measure lands as a proposal in your list, with the cause and the measured value.

5

Approval

Nothing is changed until you approve. Decisions about a theme change remain a separate project anyway.

Four-way comparison

DIY / pro / agency / automated

CriterionDo it yourselfSEO pro (manual)AgencyHow I do it
Costs€0 – PageSpeed Insights is freehourly rate, €100 with memonthly retainer, depending on contract€99–399 per month, depending on module scope, plus credit usage as consumed
Your time investmenthigh: measuring, interpreting, implementinglow, after briefinglow, after coordinationapprovals, nothing else
Measurement rhythmwhen you rememberoccasional, per assignmentdepending on contractcontinuous, every page
Expertise neededgetting into metrics and causesthey bring itthey bring itreport in plain language, interpretation included
Comparability over timescreenshots instead of a measurement seriesonly with clean documentationdepends on the reportinga continuous measurement series per page
Implementationyou yourselfper quotevia ticketsproposal, your approval, then implementation
To be honest: for a one-off check you don’t need me – PageSpeed Insights delivers a first finding for free in two minutes. And if a relaunch or theme change is coming up, the work belongs in a manually run project with a pro or an agency, not in a monthly module.
What’s in it for you

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:

01

A fixed standard

100/100/100/100 and green Core Web Vitals, mobile and desktop – not “it works, doesn’t it”, but a number

02

Fewer bounces

Fast pages keep mobile visitors in the purchase process to the end

03

Comparability

Every measurement lands in the series – you see what an update did to the values

04

Clear responsibilities

The report separates what’s down to your own system from what’s down to the theme or third parties

Frequently asked questions

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.

Context

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