Why Every ZKO Website Ships With an Actual Performance Budget

Every website we ship carries a performance budget, agreed with the client before a single line of front end code gets written, not measured for the first time once the site is live and something is already slow. It is a short document: a maximum page weight, a target Largest Contentful Paint, a hard limit on third party scripts, all specific numbers rather than a vague promise to make it fast.

Gun Trader is the clearest example we have. Rebuilding the UK's biggest gun marketplace meant thousands of listing pages, each with several images, on a userbase that includes plenty of people browsing on rural mobile signal rather than fibre broadband in a city. We set a budget of under two seconds to Largest Contentful Paint on a throttled mobile connection, non negotiable, and every image pipeline and script we added had to fit inside that number rather than the number bending to fit whatever the design wanted to include.

Budgets change how design decisions get made, which is the actual point of writing one down. A designer proposing a full width video background against a two second budget gets that conversation on day one, before a client falls in love with a comp that the budget was never going to allow, rather than in a difficult call three weeks before launch. We would rather lose an argument about a video background in a design review than lose it in production, in front of a client, watching their own site load slowly on their own phone.

A performance budget is not a technical afterthought bolted onto a finished design. Treated properly, it is a design constraint agreed at the start, exactly like a brand's colour palette or its typeface, and it shapes decisions just as early, on every build we ship, not only the ones where speed was ever going to be a headline feature.