What does this service cover?
Most speed work starts in the wrong place: a test tool is opened and the resulting score is discussed. But that score is an estimate produced on one device over one connection; its relationship to what your customers experience is limited.
We start with field data: measurements collected from real visitors' devices tell you what the problem is, and lab tests tell you why.
What we look at
Instead of a single score we look at three things the user actually feels, and we measure each page type separately:
Time for the largest content to appear: the moment the user calls the page "open"
Interaction response: how quickly the page reacts when a button is pressed
Layout shift: the page jumping around as content loads
Home, collection, product and checkout separately, split by mobile and desktop
Where the biggest gains come from
We split the weight into four groups: theme code, app code, images and third-party scripts. In our experience most of the gain comes from the last two.
Cleaning up code left in the theme by apps that were removed
Consolidating several apps doing the same job
Moving images to modern formats and deferring off-screen ones
Simplifying analytics and advertising scripts
Font loading strategy
We do not recommend cosmetic changes that raise a test score without changing the real experience.
How we work
01
Measurement
Field data and lab tests are collected by page type to establish where the problem actually is.
02
Separating the sources
A loading breakdown is produced, splitting theme, app, image and third-party weight.
03
Priority list
Each item is scored by expected gain and implementation risk, and the work list follows that order.
04
Implementation
Fixes are made step by step, with a measurement at each step and each change logged with its date.
05
Verification
We wait for the field data to settle, and track conversion and add-to-cart rate alongside it.