Website Redesign: A UX Checklist Before You Touch a Single Pixel

Most website redesigns start with a visual complaint — "it looks dated" — and end up producing a site that looks newer but converts no better, because the underlying usability problems that were actually costing business were never diagnosed before the redesign began. A proper UX checklist run before any design work starts prevents this.

Audit the existing site's real behaviour data first. Analytics showing where users actually drop off, heatmaps or session recordings showing where they click and where they get stuck, and search console data showing what people are actually looking for when they land on the site all tell you what to fix. Redesigning based only on internal opinion about what "looks old" skips the evidence that would tell you what to change.

Talk to actual users or customers before finalising a new direction, even briefly. Five short conversations with real customers about what frustrates them on the current site, or what almost stopped them from buying, routinely surfaces problems that internal stakeholders never notice because they know the site too well to see it the way a first-time visitor does.

Define what "better" means in measurable terms before design begins: a specific conversion rate target, a reduction in a particular support ticket category, a faster completion time for a key task. A redesign without a measurable definition of success cannot be evaluated as successful or not — it can only be judged on taste, which is a much weaker basis for a significant investment.

Preserve what is already working. It is tempting to change everything in a redesign, but content, navigation labels, or flows that are performing well in the current analytics should be identified and deliberately preserved, not swept away by a general desire for everything to feel new.

Plan for a controlled rollout where possible — an A/B test of the new design against the old on a meaningful sample of traffic, rather than switching everyone over at once, catches problems (an unexpected drop in conversion, confusion around a new navigation pattern) while they are still cheap to fix.

Finally, plan the content and SEO migration carefully. A redesign that changes URLs, page structure, or content without careful redirects and a deliberate SEO transition plan can quietly cost a business meaningful organic search traffic — sometimes for months — even when the new design itself is a genuine improvement.

Finally, resist the pressure to launch everything at once just because a redesign has been in progress for a while. A phased release — shipping the highest-confidence improvements first while continuing to test riskier changes — consistently produces better outcomes than a single large launch where any problem is tangled up with dozens of simultaneous changes and much harder to isolate and fix.