Your visitors leave before your product finishes loading.
I profile your site on a real phone, find what costs you the paint, and ship the fixes. You get Core Web Vitals in the green and the raw reports that prove it.
What a slow site actually costs you.
You pay twice for every visitor.
Ads and content bring someone to the page, then a four second paint sends them back to the search results. You paid for that click and got nothing for it.
Google ranks the wait.
Core Web Vitals are a ranking signal. A failing largest contentful paint holds a page down no matter how good the copy on it is.
The desktop score is not the problem.
A green number on a desktop test means little when your buyer is on a mid range Android over 4G. That is the device I test on.
What I actually change.
-
The critical path
Render blocking scripts and stylesheets come out of the head, fonts get preloaded and metric matched, and the browser paints text on the first pass instead of the third.
-
The JavaScript you ship and never run
Most sites download a whole framework to animate three sections. I cut what is dead, defer what can wait, and load the rest only where it earns its bytes.
-
Media that holds the paint hostage
Hero images get sized, converted, and served at the width the layout actually asks for. Video that autoplays above the fold stops blocking the first frame.
-
Layout shift
Everything that arrives late gets its space reserved before it lands — images, embeds, and the font swap that quietly drops your headline one line down the page.
What lands in your inbox.
The same tests, run before and after, with the raw reports attached. Median mobile PageSpeed across the last five projects is 97, up from 35, and largest contentful paint went from 4.1s to 1.2s on a throttled Moto G4.
Speed questions, answered.
Most sites I take on land above 90 on mobile. What I promise before the audit is the measurement, not a number: I profile the site first, tell you what each fix is worth, and you decide how far down the list to go.
No. Speed work is plumbing — the page looks the same and arrives sooner. Anything that would change a layout goes on a staging URL for you to review before it ships.
Yes. Webflow has its own ceiling and I know where it sits: what to cut, what to move out of the head, and the point where the platform itself is the bottleneck and a migration is the honest answer.
A typical pass runs one to two weeks from audit to shipped, depending on how much of the site is custom and how quickly your review comes back.
Send me the URL.
I run the audit and reply within a day with what I would fix first and what it is worth.
Get a speed audit