Shopify image hosting origins and the Core Web Vitals
99.2% of audited sites (8,174) serve images from their own origin (97% good LCP, 1.3s median), 82.0% load images from external domains or CDNs (96% good LCP), and 10.5% inline base64 data-URI images (95% good LCP, 1.5s median).
99.2% of Shopify sites use same-origin images image hosting origin, and 97% of those sites have a good LCP, against 97% of all Shopify sites.
LCP pass rate
Sites with same-origin images image hosting origin have a good LCP most often: 97%, against 97% of all Shopify sites. Sites with inline data-URI images image origin have it least often, at 95%.
Image hosting origin usage
Same-origin images is the most used: 99.2% of Shopify sites.
LCP phase breakdown by image hosting source
Across the four LCP sub-parts in Chrome's field data, the largest gap between same-origin images and inline data-URI images sits in resource load delay: 630 ms median for same-origin images against 776 ms for inline data-URI images. Overall median field LCP is 1.3s for same-origin images (TTFB 304 ms, load delay 630 ms, load duration 153 ms, render delay 305 ms) versus 1.5s for inline data-URI images.
Image hosting origin per framework
Vue sets the image hosting origin to same-origin images on 99% of its sites, and 97% of Vue sites have a good LCP, against 97% of all sites.
Vue 933 sitesSame-origin (self-hosted) 99%Cross-origin / external CDN 93%Inline data: URI images 12%97%
jQuery 3,149 sitesSame-origin (self-hosted) 99%Cross-origin / external CDN 88%Inline data: URI images 17%95%Image hosting origin or the framework
Does the framework explain the image hosting origin difference?
On jQuery, 95% of sites with same-origin images image hosting origin have a good LCP, against 95% of jQuery sites with cross-origin images. Within jQuery, sites with same-origin images and cross-origin images have a good LCP about equally often. On Vue, 97% of sites with same-origin images image hosting origin have a good LCP, against 96% of Vue sites with cross-origin images. Within Vue, sites with same-origin images and cross-origin images have a good LCP about equally often.
Why base64 inline data-URI images bloat HTML and delay LCP
Serving images from the same origin (99.2% of sites, 8,174) reuses the primary HTTP/2 or HTTP/3 connection (97% good LCP, 1.3s median). Cross-origin images (82.0%, 96% good LCP) include both dedicated media CDNs and external third-party pixels. Inlining images as base64 data: URIs (10.5% of sites, 865) inflates HTML/CSS byte size by ~33% and blocks the parser, dropping good LCP to 95% (1.5s median).
Understanding image hosting origins
Serve images from your primary origin to share the main connection, or add <link rel="preconnect"> if you use a dedicated cross-origin image CDN, and avoid base64 data: URIs for anything larger than tiny CSS icons. Across all audited sites, 99.2% use same-origin images (97% good LCP) while 10.5% inline data: URIs (95% good LCP).
By Arjen Karel · 8,246 crawled Shopify sites (3,818 with desktop Chrome field data, 8,240 sites with image hosting origin, October 2026). How we measured.