Nuvemshop lifted good LCP from 57% to 96% by fixing image loading

Nuvemshop is an ecommerce platform whose merchants build storefronts from movable sections such as carousels, banners and product grids. In a case study published on web.dev, its team describes how it fixed image loading across every theme and reports the result: the share of its stores with a good Largest Contentful Paint (LCP) went from 57% to 96% in one year. This is Nuvemshop’s own account, published by Google’s web.dev, not an independent test. It belongs with the other technical SEO case studies on this site.

What was wrong with Nuvemshop’s image loading?#

At the start of 2025, only 48% of Nuvemshop stores passed Core Web Vitals thresholds and 57% had healthy LCP scores. The team’s first hypothesis was image weight or server latency, and it says that hypothesis “was wrong.” Analysis across thousands of stores showed three root causes: CSS transitions delayed when elements counted as visible, so the browser sometimes chose a banner in the second section as the LCP element; lazy loading sat on images at the top of the viewport that needed to load at once; and the most important images lacked priority signals.

Which fixes did Nuvemshop ship?#

The team applied the fixes across all core themes and page types. It removed CSS transitions from first-position sections, removed loading="lazy" from the first image in first-position sections, and added fetchpriority="high" to the likely LCP image. It also added validation so the priority hint applies only where an element could actually be the LCP candidate, because marking too many images high priority defeats the purpose. The write-up also lists edge caching, applied carefully to avoid stale pricing and inventory.

What results did Nuvemshop report?#

LCP good rose from 57% to 96%, and the overall Core Web Vitals pass rate from 48% to 72%. For mobile visitors from Google organic search, the case study reports an 8.9% increase in conversion rate (session to paid order) and an 8.4% increase in cart engagement (session to cart). It says mobile drove the biggest gains.

How should this case study be read?#

The conversion and cart figures are percentage changes with no published baseline rates, so they are recorded here as changes from a zero baseline. The case study also bundles several changes at once, so it does not isolate the effect of removing lazy loading alone. The transferable point is narrower: check which element the browser actually reports as LCP, and never lazy-load it.