The real cost of a slow website
Every 100 milliseconds of latency has a price. We break down the revenue math behind Core Web Vitals.
"It feels fast enough on my machine" is the most expensive sentence in web development. Your machine is a high-end laptop on office wifi. Your users are on a mid-range phone with a patchy connection, and they are far less patient than you are.
The revenue math
The relationship between speed and money is not subtle. Study after study — and every conversion dashboard we have ever audited — shows the same shape: as load time climbs, conversion falls, and it falls fast. A site that takes three seconds to become usable is leaving a meaningful share of its revenue on the table compared to one that takes one.
The mechanism is simple: every extra second is another moment for a distracted user to give up. Bounce compounds. Slower pages mean fewer page views per session, lower conversion per visit and less repeat traffic.
Google is grading you
Since Core Web Vitals became a ranking signal, performance is no longer only a conversion issue — it is a traffic issue. Largest Contentful Paint (LCP), Interaction to Next Paint (INP) and Cumulative Layout Shift (CLS) are measured on real visits and feed directly into search ranking. A slow site loses at both ends of the funnel: fewer visitors arrive, and fewer of the ones who do convert.
Where the wins actually are
After auditing dozens of sites, the culprits are boringly consistent:
- Images. Unoptimised, wrongly sized, no modern format. This is almost always the single biggest LCP offender.
- Fonts. Web fonts that block rendering or cause layout shift. Self-host, preload, and set
font-displaysensibly. - Render-blocking scripts. Third-party tags, analytics and chat widgets loaded eagerly in the head.
- Layout shift. Images and embeds without reserved dimensions, causing the page to jump as it loads.
None of this is exotic. The reason sites stay slow is not that the fixes are hard; it is that nobody owns the number.
Measure real users
A perfect lab score means nothing if your actual visitors are having a bad time. Prioritise field data — real-user monitoring from actual devices and networks — over synthetic lab runs. The lab tells you what could happen; the field tells you what is happening to the people who pay you.
Speed is not a nice-to-have you get to after the redesign. It is the redesign. Fix the images, tame the fonts, defer the scripts, reserve your layout — and watch both your ranking and your conversion rate move in the same, profitable direction.
Frequently asked questions
What is a "good" load time in 2026?
Aim for a Largest Contentful Paint under 2.5 seconds and Interaction to Next Paint under 200 milliseconds on real mobile devices. Faster is always better, but crossing those thresholds is where you stop losing users and ranking to speed.
Will a faster site really increase revenue?
In our experience, yes — consistently. Conversion is highly sensitive to load time, and because Core Web Vitals also affect search ranking, speed improvements compound: more visitors arrive and more of them convert.
What is the single highest-impact fix?
For most sites it is images — serving correctly sized, modern-format, lazy-loaded images. It is usually the largest element on the page and the biggest driver of Largest Contentful Paint.
Keep exploring
Enjoyed this? We build the things we write about.
From first sketch to scaled platform — let's talk about your project.
Start a project
