The Page Speed To Revenue Link, Quantified
100ms shaved off LCP is worth roughly 1 percent in conversion. The maths over a year.
Page speed is usually argued as a ranking factor. That undersells it. The larger effect is on conversion, and conversion is where the money is.
What the delay actually costs
Every additional second before a page becomes usable loses a share of the people waiting. The effect is steepest at the start, so the difference between a fast page and a slightly slow one is larger than the difference between slow and slower.
On mobile, on real connections, that gap is wider than it looks in a test on office wifi.
The metric to watch
Largest Contentful Paint is the one that correlates most directly with the visitor's experience, because it approximates the moment the page looks usable.
Improving it is usually not a rewrite. It is nearly always the same handful of causes.
The four usual culprits
Images. Oversized, uncompressed, in the wrong format, and not sized for the viewport. This is the single most common cause and the easiest to fix.
Fonts. Custom fonts that block rendering, or that swap late and shift the layout. Preloading the ones you need and limiting the number of weights removes most of it.
Scripts. Analytics, chat widgets, heat maps, tag managers, each added at some point for a reason nobody remembers. Every one costs load time on every visit.
Server response. If the server takes a long time to send the first byte, nothing downstream can recover it.
Do the maths on your own numbers
Take your monthly traffic, your conversion rate, and your average customer value. Model a modest improvement in conversion from a faster page.
For most businesses that number is larger than the cost of the engineering, and it recurs every month rather than being a one-off.
That is the argument that gets speed work prioritised, and it is an argument in revenue rather than in milliseconds, which is why it works with people who do not care about milliseconds.
Measure the real thing
Lab tests run on a fast connection from a nearby server. Field data reflects what your actual visitors experience on their actual devices.
Optimise against the field data. A perfect lab score with poor field performance means you have optimised for the test.
Where it compounds
Speed feeds everything else. Faster pages convert better, get crawled more thoroughly, and hold attention longer, which improves the engagement signals that feed ranking.
It is unglamorous work with a return that shows up across every other channel you run.
Written by David Eid. Published .
Read next.
Contact the Ignis Team
Send through your details and we will audit your business before we reply.