We need to talk about automation. Recently, the WordPress Core Performance team has been pushing hard to automate LCP (Largest Contentful Paint) optimizations. It sounds great on paper, but in practice, it’s a minefield. The latest WordPress Performance Optimization chat revealed exactly why trying to be “smart” with browser hints like fetchpriority can backfire if the logic isn’t airtight.
The fetchpriority=”high” Paradox in Core
The main topic of the March 10th chat was Ticket #64823. This ticket highlights a specific “gotcha” where fetchpriority="high" was being incorrectly applied to images hidden inside Gutenberg blocks, such as navigation overlays. When you tell a browser an image is high priority, but that image isn’t even visible in the initial viewport, you’re essentially stealing bandwidth from the actual LCP element.
I’ve seen this happen on client sites where a hero image competes with a “hidden” mobile menu image that Core tagged as high priority. The result? A degraded LCP score and a confused developer. Consequently, a new core patch is being reviewed to refine the wp_get_loading_optimization_attributes() and wp_maybe_add_fetchpriority_high_attr() functions.
// The naive approach: Add high priority to the first image found
// The fixed approach: Ensure the image isn't hidden by block logic
function bbioon_refine_priority_logic( $attrs, $tag_name, $context ) {
if ( 'IMG' === $tag_name && isset( $attrs['fetchpriority'] ) ) {
// Additional checks for block-specific hidden states
if ( bbioon_is_image_hidden_in_block( $context ) ) {
$attrs['fetchpriority'] = 'auto'; // Downgrade to default
}
}
return $attrs;
}
add_filter( 'wp_get_loading_optimization_attributes', 'bbioon_refine_priority_logic', 10, 3 );Why the Web Worker Experiment is Sunsetting
Furthermore, there’s been movement on the Performance Lab plugin. The team is sunsetting the “Web Worker Offloading” feature. If you aren’t familiar, this was an attempt to use Partytown to offload heavy third-party scripts (like Google Analytics) to a web worker, freeing up the main thread.
While the theory is sound, the reality is that offloading scripts often breaks things in ways that are nearly impossible to debug without a deep dive into the partytown source code. Weston Ruter noted that despite 6,000+ updates to the version containing the sunset warning, there’s been zero user feedback. This suggests that either users aren’t utilizing it or they’ve already moved on to more stable WordPress Performance Optimization techniques.
If you’re still looking for ways to handle query-heavy sites, you might want to check out my previous thoughts on Critical WordPress Performance Updates: The WP_Query Shift, where we discuss backend bottlenecks that fetchpriority can’t solve.
Look, if this WordPress Performance Optimization stuff is eating up your dev hours, let me handle it. I’ve been wrestling with WordPress since the 4.x days.
Final Takeaway
Performance optimization in 2026 isn’t just about adding tags; it’s about context. Specifically, we need to ensure that Core features don’t blindly apply “optimizations” that conflict with modern block architecture. With Weston Ruter traveling for the next two weeks, expect a slight lull in Core Performance updates, but keep an eye on PR #11196 for the fetchpriority fix. Ship it carefully.
