WordPress LCP image origin (first-party vs third-party) and the Core Web Vitals
Serving the hero image from your own domain versus an external image CDN: how a separate connection adds 43 ms of resource load delay (333 ms vs 290 ms) and when <link rel="preconnect"> pays it back.
94.5% of WordPress sites use first-party LCP image hosting, and 75% of those sites have a good LCP, against 76% of all WordPress sites.
LCP pass rate
Sites with first-party LCP image hosting have a good LCP most often: 75%, against 76% of all WordPress sites. Sites with third-party image hosting have it least often, at 73%.
LCP image hosting usage
First-party is the most used: 94.5% of WordPress sites.
Where third-party LCP images lose and win time: the four LCP phases
Across the four LCP sub-parts in Chrome's field data, the largest gap between first-party and third-party sits in time to first byte (TTFB): 1010 ms median for first-party against 1067 ms for third-party. Overall median field LCP is 1.8s for first-party (TTFB 1010 ms, load delay 290 ms, load duration 131 ms, render delay 218 ms) versus 1.9s for third-party.
LCP image origin combined with preconnect and priority hints
Combining first-party with Preload reaches 80% good LCP across 568 sites (188 ms median load delay). At the other end, third-party paired with Preconnect sits at 69% across 257 sites (381 ms median load delay).
| Image origin | No priority hint | <link rel="preconnect"> | fetchpriority="high" | <link rel="preload"> |
|---|---|---|---|---|
| Same-origin / first-party | 74% 37,616 sites | 86% 48 sites | 79% 6,009 sites | 80% 1,715 sites |
| Third-party domain | 74% 1,409 sites | 69% 709 sites | 80% 413 sites | 71% 109 sites |
LCP image hosting per framework
Vue sets the LCP image hosting to first-party on 97% of its sites, and 75% of Vue sites have a good LCP, against 76% of all sites.
jQuery 45,936 sitesSame-origin / first-party 95%75%
Vue 3,830 sitesSame-origin / first-party 97%75%
React 1,732 sitesSame-origin / first-party 93%66%LCP image origin or the framework
Does the framework explain the LCP image hosting difference?
Almost every site with first-party LCP image hosting is a jQuery site, and almost every jQuery site has first-party LCP image hosting, so the data can't tell first-party LCP image hosting and jQuery apart. On React, 64% of sites with first-party LCP image hosting have a good LCP, against 70% of React sites with third-party. Within React, sites with first-party and third-party have a good LCP about equally often.
Why third-party hero images pay a connection tax in resource load delay
When the LCP image lives on the same origin as the HTML document, the browser reuses the open HTTP/2 or HTTP/3 connection immediately. When the image sits on a separate domain, Chrome must resolve DNS, complete a TCP or QUIC handshake, and negotiate TLS before requesting the first image byte. That handshake shows up directly in Chrome's field data: third-party LCP images have a 333 ms median resource load delay compared with 290 ms for first-party images (43 ms slower to start), even though dedicated image CDNs finish the download slightly faster (126 ms vs 131 ms).
Understanding LCP image origin
Serve your hero image from your own domain (behind a full-site CDN) or add <link rel="preconnect" crossorigin> to your external media domain. On WordPress, 94.5% of sites serve the LCP image from their own domain (75% good LCP) and 5.5% load it from an external CDN or offload domain (73% good LCP).
By Arjen Karel · 63,295 crawled WordPress sites (17,005 with desktop Chrome field data, 48,028 sites with LCP image hosting, October 2026). How we measured.