Open Graph Debugger

Open Graph checker: what the report shows

When someone pastes a link into Facebook, LinkedIn, X, Telegram, Slack or a messenger, the platform's crawler downloads the page and reads the Open Graph tags in its <head>. Those tags decide the title, description and picture of the preview card. This checker fetches your page the same way, from our server and without running JavaScript, and shows three things:

• Preview cards for Facebook, X and LinkedIn, built from the tags it found. They follow each network's layout closely, but they are an approximation: the real card depends on the network's current design and its cache.
• A tag list with seven Open Graph tags (og:title, og:description, og:image, og:url, og:type, og:site_name, og:locale) and four X tags (twitter:card, twitter:title, twitter:description, twitter:image). Each is marked present or missing.
• A score from 0 to 100. Each tag has a weight (title and image count most, locale least), and the score is the share of weight found. It measures completeness, not quality: a present but badly written title still counts.

Under the previews you also see the real pixel size of og:image as your browser loaded it, the size declared in og:image:width and og:image:height, the og:image:alt text and a warning if the image URL is relative.

Which Open Graph tags are required

The Open Graph protocol (ogp.me) names four required properties. Add the description and the X card type and you have a set that works everywhere:

A minimal head looks like this:

<meta property="og:title" content="Spring sale: 30% off running shoes">
<meta property="og:description" content="Two weeks only, free returns.">
<meta property="og:type" content="website">
<meta property="og:url" content="https://example.com/sale">
<meta property="og:image" content="https://example.com/img/sale-1200x630.jpg">
<meta property="og:image:width" content="1200">
<meta property="og:image:height" content="630">
<meta property="og:image:alt" content="Three pairs of running shoes on a track">
<meta name="twitter:card" content="summary_large_image">

OG image size

One image at 1200×630 pixels (ratio 1.91:1) covers Facebook, LinkedIn and X large cards. Keep the key text and logo in the centre, because some apps crop the edges or show a square thumbnail.

Declaring og:image:width and og:image:height lets Facebook render the card on the first share instead of waiting for the image to download. Compress the file: a 3 MB banner slows the crawler and some messengers skip heavy images.

Why the preview did not update, and how to refresh it

Every network caches what it scraped. After you fix the tags, this checker shows the new version straight away, but the network keeps the old card until it re-scrapes. Force it with the official tools:

• Facebook, Instagram links, Messenger: open the Sharing Debugger on developers.facebook.com, paste the URL and press "Scrape Again". Without that, Facebook refreshes roughly every 30 days.
• LinkedIn: run the URL through the Post Inspector at linkedin.com/post-inspector. Inspecting the URL fetches it again; otherwise LinkedIn keeps the cache for about seven days.
• Telegram: send the link to the official bot @WebpageBot. It updates the preview that Telegram stores for that URL.
• X: the old Card Validator no longer shows previews, and there is no manual refresh. The composer shows the card when you paste the link. If X keeps an outdated card, share an address with a harmless query parameter such as ?v=2, which X treats as a new URL.

Images are cached by URL too. If you replace the picture under the same file name, some apps keep showing the old one. Upload the new image under a new name and update og:image.

X cards: what falls back to Open Graph

X reads its own twitter:* tags first. If twitter:title, twitter:description or twitter:image are missing, X uses og:title, og:description and og:image. So you do not need to duplicate everything: twitter:card is the one tag worth adding to a page that already has full Open Graph. Our X preview follows the same fallback. Note that X tags are normally written with the name attribute; this checker reads them from name only.

Common Open Graph mistakes

1. Relative image path. /img/cover.jpg works in a browser but not in a crawler. Use the full https:// address; the report flags relative paths.
2. Tags added by JavaScript. Social crawlers do not run scripts. If your framework sets meta tags on the client, the crawler sees the empty template. Render them on the server or prerender the page. This checker also does not run JavaScript, so a missing tag here usually means a missing tag for Facebook.
3. Crawler blocked. A Disallow in robots.txt, a firewall "bot fight" mode, geo-blocking or a login wall make the network see an error page. Check robots.txt and the server response in HTTP headers.
4. One template for every page. The same title and picture on all URLs make shares look like the home page. Generate og:title, og:description and og:image per page.
5. Wrong og:url. Facebook treats og:url as the canonical address and builds the card from that page. If every page points to the home page, every share shows the home page.
6. Several og:image tags by accident. Plugins and themes sometimes both output one. Usually the first is used, and it may not be the one you want.
7. Redirect chains. Crawlers follow redirects, but long chains or a redirect to a login page break previews. Check them with the redirect checker.

Open Graph and SEO

Google does not rank pages by Open Graph tags, and it builds search snippets from <title> and the meta description, not from og:*. Open Graph matters for clicks from social networks and messengers, where a card with a clear picture gets noticed. Check the search side separately: the meta tag checker covers title, description and canonical, and the SERP preview shows how the page looks in Google.