The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
If you embed a video or a map in a WordPress post, and the embed is not the very first thing in that post, it stays empty for a while. Readers reach it before it arrives.
WordPress adds a short delay instruction to the pictures and embedded frames inside a post. The instruction means: do not load me until the visitor scrolls close. For media deep in a post, that works well. The page loads fast, and the video is ready by the time a reader gets there.
WordPress makes one exception. The first picture or frame in a post does not get the delay, because it is assumed to be the first thing people see. That assumption holds for single-column posts, where the layout stacks one item after another.
It breaks when a post starts with two elements that compete for the top. An intro video above a main video. A header image above a product video. Then the second element is the one a visitor sees from the start, and the second element is the one carrying the delay. The reader looks at an empty box.
The change is a filter with one search line. It tells WordPress to load a certain embed right away instead of leaving it delayed.
add_filter( 'wp_iframe_tag_add_loading_attr', function( $value, $iframe, $context ) {
if ( 'the_content' === $context ) {
if ( false !== strpos( $iframe, 'youtube.com' ) ) {
return 'eager';
}
}
return $value;
}, 10, 3 );
Put the name of the platform you embed in the search line: vimeo.com, maps.google.com, spotify.com. Only embeds that match change. Everything else keeps the delay.
A quick way to see whether your site has the problem: open one of your posts in a private window, press F12, and watch the network list. A delayed embed starts fetching long after the page has finished drawing.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Just optimized a WordPress site: 8.2s -> 1.1s LCP, 96 PageSpeed mobile.
Stack: WP Rocket + Cloudflare APO, WebP + lazy load, removed unused JS/CSS.
Result: 86.6% faster, all Core Web Vitals pass.
If your WordPress is slow, comment your URL - I'll do a free 2-min Loom teardown for first 3 people
Client
A baseball sports medicine organization needed a professional WordPress website for its 2026 conference, designed to communicate the event's value, showcase its faculty, and make registration and conference information easy to access.
Challenge
The website needed to bring together a large amount of conference information, including the program schedule, faculty, attendance options, venue details, sponsors, FAQs, and registration, while maintaining a strong baseball sports medicine identity and making the information easy to navigate.
Solution
I designed and developed a custom WordPress website with a bold baseball-inspired visual direction, structured program sections, faculty profiles, conference experience options, venue information, sponsor display, FAQs, and strategic registration CTAs. The responsive layout was built to keep important conference details accessible across desktop and mobile.
Expected Result
The website is designed to strengthen the conference's online presence, communicate its professional value, simplify registration and event planning, and give physicians, athletes, coaches, and sports medicine professionals a clear path to discovering and attending the program.
Keywords
I’m building my first fully AI-made, YouTube video end-to-end, from idea to publish. Now I have the first usable clip.
Not yet the full YouTube video.
But enough to prove the direction.
The reason I started this is simple:
I want YouTube to become a deeper authority channel for my work in AI.
YouTube hits different.
If someone sits with you for 10 minutes, the relationship is different.
The depth is different.
The problem is that I don’t like being on camera. I really don't enjoy it.
So instead of forcing myself into a content workflow, I’m building the system I would want to use.
A fully AI-made YouTube workflow:
Idea.
Script.
Avatar.
Visuals.
Editing.
Rendering.
Publishing.
The thinking is mine,
the taste and judgement are mine,
the delivery is the avatar.
AI doesn't remove all the work.
It moves it into:
judgment.
workflow design.
quality control.
That is the real thesis I’m testing.
Not whether AI can make a video.
It can, that's been proven, now moving to mastering long form.
I’ll keep documenting the build as I finish the full video.