Why website speed matters more for conversion than design

A beautiful design that takes four seconds to load loses to an ugly design that loads in one. It sounds harsh, but data has confirmed it for years: every extra second of loading means a measurable drop in conversions. The problem is that "speed" is a vague word - Core Web Vitals break it down into three concrete, measurable things.

LCP: when the main content is visible

Largest Contentful Paint measures how long it takes for the largest visible element to appear on screen - typically the main image or the heading. Google recommends under 2.5 seconds. Above that threshold a website feels slow, even if the user can't put a name to it - they just leave.

CLS: why the website "jumps"

Cumulative Layout Shift measures how much the content shifts unexpectedly during loading. The classic scenario: the "Buy" button appears, the user clicks it, but in the meantime an image at the top has finished rendering and the button jumped a bit lower - the click hits something else. In an online store this is a direct loss of an order, not a cosmetic flaw.

What can actually be done about it

On bazarfoto.cz it was a chain of specific decisions, not one trick: the LCP image got loading priority, static files have permanent caching with versioning, and pages are rendered by the server without a heavy extra JS framework. The result is measurable - Lighthouse Performance 100 on desktop, 90–100 on mobile, not a guess off the cuff.

Where it ends

Website speed isn't a project you do once and forget. It's a consequence of architectural decisions made at the start - what is rendered where, how much JavaScript the website sends to the client, how the images are optimized. That is exactly what a conversion rate audit covers: finding where exactly the website loses seconds, and calculating what that costs.