Measure
Field data if available (CrUX), lab data (Lighthouse) and a real-device trace. We record LCP, INP, CLS, total page weight and the number of third-party origins.
Website speed
Website speed, measured before and after
A slow site is usually not a hosting problem. It is a pile of third-party scripts, an unoptimised font stack and a page builder doing more than anyone asked. We find the weight, remove it, and show you the difference.
Page speed is a user-experience metric before it is a ranking factor. A visitor on a phone, on a train, with two bars of signal does not care that the server responded in 80 milliseconds if the page waits four seconds for a chat widget, a heatmap and a consent manager to load.
We measure on a throttled mobile connection, not a developer's laptop. We record the baseline, identify what is actually blocking the main thread, and remove or defer it. Then we measure again, under the same conditions, and give you the numbers — including the ones that did not improve.
Method
Four passes over the same page
Remove
Every script earns its place or goes. Chat widgets become a link until requested. Heatmaps and session recorders are removed. Fonts are subset and self-hosted. Images are sized and served in a modern format.
Fix
Render-blocking CSS, layout shift from unsized images, long tasks from hydration, and server response time. The fixes are made in the codebase, not with a plugin that re-adds weight elsewhere.
Monitor
A performance budget in the build, a weekly check, and an alert when a new release pushes a metric over the line. A one-off optimisation decays within months without this.
Targets
What the numbers mean
| Metric | Measures | Good | Needs work |
|---|---|---|---|
| LCP | How long until the main content is visible | 2.5s or less | Over 4s |
| INP | How quickly the page responds to a tap | 200ms or less | Over 500ms |
| CLS | How much the layout jumps while loading | 0.1 or less | Over 0.25 |
| Page weight | Bytes transferred for the first view | Under 1MB | Over 2.5MB |
| Third-party origins | External scripts, fonts and trackers | Zero to two | Five or more |
On the phrase “loads in under a second”
On the phrase “loads in under a second”
- Read how this site is builtOur own site ships one CSS file, one small script and no third-party requests. We publish the method, not just the claim.
- Speed checklistA practical list you can work through without hiring anyone.
Questions
The things people ask first
Will speed work break the design?
It should not. Most of the weight is invisible: scripts, fonts, images and redirects. Where a visual element is genuinely costly, we will show you the cost and offer a lighter version rather than silently removing it.
Is a CDN enough?
A CDN helps with distance and caching, but it cannot fix a 3MB page of JavaScript. Fix the weight first, then add infrastructure if the audience is global.
Can you optimise a WordPress or Wix site?
Partially. We can remove plugins, defer scripts, compress images and improve hosting, but some platform weight cannot be removed. If the ceiling is too low, we will say so and quote a rebuild.
How do you measure?
Lighthouse on a throttled mobile profile, PageSpeed Insights field data where available, and a real-device trace. We record the date, the profile and the URL with every number.
Related
Let’s find out whether technology is the answer
Tell us what you are trying to achieve. We will tell you honestly whether we can help, what it would take, and roughly what it would cost — before anyone signs anything.