Script initiators and the Core Web Vitals
97.7% of audited sites (192,444) load scripts directly from the HTML parser (95% good INP, 86ms median), while 75.0% inject scripts via other scripts (95% good INP, 88ms median) and 18.2% trigger scripts from other initiators (93% good INP).
75.0% of sites use script-injected scripts script initiator type, and 95% of those sites have a good INP, against 95% of all sites.
18.2% of sites use other script initiators script initiator type, and 84% of those sites have a good LCP, against 82% of all sites.
Core Web Vitals pass rate
Sites with script-injected scripts script initiator type have a good INP most often: 95%, against 95% of all sites. Sites with other script initiators script initiators have it least often, at 93%.
Sites with other script initiators script initiator type have a good LCP most often: 84%, against 82% of all sites. Sites with script-injected scripts script initiators have it least often, at 81%.
Script initiator type usage
Parser-initiated scripts is the most used: 97.7% of sites. Script-injected scripts is 75.0% of sites.
Understanding script initiators
Parser-discovered scripts versus cascading script-injected chains
When scripts are declared in HTML (97.7% of sites, 192,444), the browser preload scanner discovers them immediately (95% good INP, 86ms median). On 75.0% of sites (147,825), scripts are dynamically injected by other scripts such as Google Tag Manager or loader snippets (95% good INP, 88ms median), while 18.2% (35,826) trigger scripts via other initiators (93% good INP, 89ms median).
Prefer declarative <script defer> tags in HTML over JavaScript loader chains that inject second- and third-level scripts at runtime, and throttle tag-manager injection so late scripts do not fire during user input. Across all audited sites, 97.7% use parser-initiated scripts (95% good INP) and 75.0% use script-injected loaders (95% good INP).
How script initiator type plays out per platform
What the field data shows when you isolate a single CMS or framework.
87.1% of WordPress sites load script-injected scripts, while 10.4% with other initiators pass INP at 97%.
Across 65,494 WordPress sites, 98.7% load scripts from the HTML parser (98% good INP, 86ms median), 87.1% inject scripts from other scripts (98% good INP), and 10.4% use other initiators (97% good INP, 88ms median).
99.9% of Shopify stores use script-injected loaders, while 19.1% with other initiators pass INP at 94%.
Across Shopify storefronts, 99.8% load parser-initiated scripts (96% good INP), 99.9% load script-initiated scripts (96% good INP, 97ms median), and 19.1% have other script initiators (94% good INP, 104ms median).
99.6% of React & Next.js sites declare scripts in HTML and 95.1% also load script-initiated chunks (95% good INP).
Across 9,570 React sites, 99.6% use parser-discovered <script> tags (95% good INP), 95.1% load script-initiated chunks (95% good INP), and 71.6% use other initiators (95% good INP).
By Arjen Karel · 210,902 crawled sites (196,263 with all devices Chrome field data, 196,999 sites with script initiator type, October 2026). How we measured.