LCP image byte weight on React sites
48.5% of audited sites (3,110) keep their LCP hero image under 40 KB and pass LCP at 89% (1.3s median), whereas the 9.3% shipping hero images over 300 KB drop to 62% (2.2s median).
48.5% of React sites use an LCP image under 40 KB LCP image byte weight, and 89% of those sites have a good LCP, against 85% of all React sites.
LCP pass rate
Sites with an LCP image under 40 KB LCP image byte weight have a good LCP most often: 89%, against 85% of all React sites. Sites with an LCP image over 300 KB hero image size have it least often, at 62%.
LCP image byte weight usage
An LCP image under 40 KB is the most used: 48.5% of React sites.
The four LCP phases
Across the four LCP sub-parts in Chrome's field data, the largest gap between an LCP image under 40 KB and an LCP image over 300 KB sits in time to first byte (TTFB): 467 ms median for an LCP image under 40 KB against 1183 ms for an LCP image over 300 KB. Overall median field LCP is 1.3s for an LCP image under 40 KB (TTFB 467 ms, load delay 439 ms, load duration 131 ms, render delay 161 ms) versus 2.1s for an LCP image over 300 KB.
LCP image byte weight per platform
Wix sets the LCP image byte weight to an LCP image under 40 KB on 61% of its sites, and 94% of Wix sites have a good LCP, against 85% of all sites.
LCP image byte weight or the platform
Does the platform explain the LCP image byte weight difference?
On Wix, 94% of sites with an LCP image under 40 KB LCP image byte weight have a good LCP, against 96% of Wix sites with an LCP image between 40 and 100 KB. Within Wix, sites with an LCP image under 40 KB have a good LCP less often. On WordPress, 62% of sites with an LCP image under 40 KB LCP image byte weight have a good LCP, against 66% of WordPress sites with an LCP image between 40 and 100 KB. Within WordPress, sites with an LCP image under 40 KB and an LCP image between 40 and 100 KB have a good LCP about equally often.
How intrinsic versus rendered megapixels waste hero bandwidth
The median LCP image is 0.23 intrinsic megapixels rendered into a 0.12 megapixel slot. Serving unresized desktop banners to mobile viewports inflates hero transfer bytes and extends LCP load duration. Across 3,110 sites (48.5%) with a hero image under 40 KB, 89% pass field LCP (1.3s median), dropping to 62% (2.2s median) on the 9.3% (597) exceeding 300 KB.
Understanding LCP image byte weight
Compress hero images below 100 KB using AVIF or WebP and serve responsive srcset widths matched to the rendered viewport slot. Across all audited sites, 48.5% keep the hero image under 40 KB (89% good LCP, 1.3s median) versus 9.3% above 300 KB (62% good LCP, 2.2s median).
By Arjen Karel · 9,570 crawled React sites (9,110 with all devices Chrome field data, 6,415 sites with LCP image byte weight, October 2026). How we measured.