HTML document size and the Core Web Vitals
35.3% of audited sites (73,312) ship an initial HTML document under 14 KB and reach 88% good LCP (1.2s median), compared to 83% good LCP (1.5s median) on the 25.4% shipping 30 to 75 KB of HTML.
28.2% of sites use 14 to 30 KB of HTML initial HTML document size, and 89% of those sites have a good LCP, against 86% of all sites.
28.2% of sites use 14 to 30 KB of HTML initial HTML document size, and 99% of those sites have a good INP, against 99% of all sites.
Core Web Vitals pass rate
Sites with 14 to 30 KB of HTML initial HTML document size have a good LCP most often: 89%, against 86% of all sites. Sites with over 75 KB of HTML HTML size have it least often, at 81%.
Sites with 14 to 30 KB of HTML initial HTML document size have a good INP most often: 99%, against 99% of all sites. Sites with over 75 KB of HTML HTML size have it least often, at 98%.
Initial HTML document size usage
Under 14 KB of HTML is the most used: 35.3% of sites. 14 to 30 KB of HTML is 28.2% of sites.
LCP phase breakdown by initial HTML document size
Across the four LCP sub-parts in Chrome's field data, the largest gap between 14 to 30 KB of HTML and over 75 KB of HTML sits in resource load delay: 235 ms median for 14 to 30 KB of HTML against 417 ms for over 75 KB of HTML. Overall median field LCP is 1.3s for 14 to 30 KB of HTML (TTFB 616 ms, load delay 235 ms, load duration 112 ms, render delay 187 ms) versus 1.6s for over 75 KB of HTML.
Across the four LCP sub-parts in Chrome's field data, the largest gap between 14 to 30 KB of HTML and over 75 KB of HTML sits in resource load delay: 235 ms median for 14 to 30 KB of HTML against 417 ms for over 75 KB of HTML. Overall median field LCP is 1.3s for 14 to 30 KB of HTML (TTFB 616 ms, load delay 235 ms, load duration 112 ms, render delay 187 ms) versus 1.6s for over 75 KB of HTML.
Understanding initial HTML document size
Why the 14 KB initial TCP congestion window sets the pace for LCP discovery
A new TCP connection transfers roughly 14 KB (10 segments) in its first round-trip before waiting for a client ACK. Across 73,312 sites (35.3%) whose HTML fits under 14 KB, 88% achieve good LCP (1.2s median, 443 ms TTFB). Sites between 14 and 30 KB (28.2%) record 89% good LCP (1.3s median), while pages bloating to 30 to 75 KB (25.4%) drop to 83% good LCP (1.5s median) and pages over 75 KB (11.1%) sit at 81% good LCP (1.6s median).
Place <meta charset>, viewport, and <link rel="preload"> for the LCP image at the very top of <head> within the first 14 KB, and trim oversized inline CSS blocks or serialized SSR JSON payloads. Across all audited sites, 35.3% ship under 14 KB (88% good LCP) while 25.4% ship 30 to 75 KB (83% good LCP).
How initial HTML document size plays out per platform
What the field data shows when you isolate a single CMS or framework.
WordPress pages over 75 KB HTML (10.9%) pass LCP at 65% vs 84% under 14 KB.
Across 65,939 WordPress sites, 11.8% ship HTML under 14 KB (84% good LCP, 1.5s median), 39.5% ship 14 to 30 KB (84% good LCP), 37.8% ship 30 to 75 KB (71% good LCP), and 10.9% exceed 75 KB (65% good LCP).
89% of Shopify stores ship over 30 KB of HTML (98% good LCP at 30 to 75 KB vs 93% over 75 KB).
Across 8,634 Shopify storefronts, 63.0% transfer 30 to 75 KB of HTML (98% good LCP, 1.2s median) and 25.9% exceed 75 KB (93% good LCP, 1.5s median).
67.6% of React & Next.js sites ship over 75 KB of HTML (84% good LCP).
Across React sites, 67.6% transfer over 75 KB of initial HTML (84% good LCP) and 17.0% transfer 30 to 75 KB (67% good LCP).
By Arjen Karel · 212,308 crawled sites (70,213 with desktop Chrome field data, 207,650 sites with initial HTML document size, October 2026). How we measured.