Rebuilding an ecommerce site is usually framed as a design decision, a new look, a fresher feel, a competitor’s site that made the old one feel dated. The actual evidence for what moves conversion points somewhere much less visually exciting: load speed and checkout friction, both of which are smaller, cheaper problems than a full rebuild.
Nuvemshop’s own published case study is a useful data point here. Focusing specifically on image loading performance, without touching the overall design, they took their Largest Contentful Paint pass rate from 57 to 96 percent across a cohort of stores, and saw an 8.9 percent lift in mobile conversion and an 8.4 percent lift in cart engagement as a direct result. Rakuten 24 ran an A/B test isolating page speed improvements on a single landing page and measured a 33 percent increase in conversion rate and a 53 percent increase in revenue per visitor, again without a design overhaul. Renault analysed over 10 million visits across 33 countries and found a strong, consistent correlation between faster load times and both lower bounce rates and more completed lead forms.
None of these are redesigns. They’re targeted fixes to how fast a page becomes usable, which Google’s own Core Web Vitals framework quantifies with two thresholds worth knowing: Largest Contentful Paint should land under 2.5 seconds, and Interaction to Next Paint under 200 milliseconds. Sites that clear both bars consistently outperform ones that don’t, in case study after case study, regardless of how the site looks.
Checkout friction, covered in more detail elsewhere in this series, is the other half of the story. Baymard Institute’s long running research on cart abandonment points to the same conclusion from a different angle: the biggest lever is whether a shopper hits a surprise cost or an unnecessary account creation step on the way to paying, well ahead of how the page looks.
Put those together and a pattern emerges that doesn’t flatter the rebuild pitch. A full redesign is a large, expensive bet that a new look will convert better, with genuinely mixed evidence that it reliably does. Fixing measurable, specific friction, a slow product page, a checkout that hides the total, is a much smaller bet with case studies showing real, repeatable returns.
That doesn’t mean a rebuild is never the right call. Sites built on genuinely outdated infrastructure, or ones that can’t be extended without fighting the platform, have real reasons to start fresh. But if the question is purely which investment predicts a return, the honest answer from the available data is to fix what’s already measurably broken before paying for something new to be built on top of it. A rebuild is a bet. Fixing what’s already measurably broken isn’t.
A rebuild is a bet. Fixing what's already measurably broken isn't.
