Free Meta Tag Checker

One check fetches the page once, as our checker, and reads the HTML the server sends. Then it fetches the share image once, reading only the start of the file, which is enough to confirm it is an image and to read its width and height. Nothing is guessed: a tag we did not find, an image we could not load or a size we could not read is reported as exactly that.

• The title and meta description, with their length in characters and their estimated width in pixels, and whether a page has none, an empty one, or more than one.
• A Google result preview on desktop and on mobile, cut where Google would usually cut it.
• Every Open Graph and Twitter tag on the page, with link card previews for Facebook and LinkedIn and for X.
• The share image's status, file type and size, checked against the size every network shows well.
• Robots, viewport, charset, language, favicon and hreflang tags, with obvious hreflang mistakes flagged.

The checker reads the raw HTML and does not run JavaScript, which is also how most social sites read a page when someone shares it. If your tags are added by a script after the page loads, they will be missing here, and probably missing on shared links too.

Google does not cut a meta description at a fixed number of characters. It cuts by width: on desktop, a snippet runs to about two lines, close to 960 pixels of 14 pixel Arial. Narrow letters such as i and l fit more characters into that space than wide ones such as m and w, which is why a 150 character description can fit on one page and be cut on another.

The checker measures the description with Arial's own letter widths and flags it when it is likely to be cut. It also flags a description under 70 characters, which gives Google little to show, a page with more than one description tag, and an empty one. Without a description, Google writes the snippet from the page text, which may not be the reason you want people to click.

Google may rewrite any snippet to match a search, so a good description is a strong suggestion, not a guarantee. Write one or two sentences that say what the page offers and why it is worth the click, and put the key point first, so it survives a cut.

The title tag is the headline of the search result and the label on the browser tab. On desktop, Google usually cuts titles wider than about 600 pixels of 20 pixel Arial and adds an ellipsis. The checker estimates the width the same way it does for descriptions, and its preview shows the title cut where Google would usually cut it.

It flags a missing or empty title, a title under 30 characters, and a page with two or more title tags. A second title usually comes from a theme and a plugin both writing one; browsers and search engines use the first, so the second only causes confusion. A title inside an inline SVG or a template does not count, because it is not the page's title.

On mobile, results wrap titles over more lines instead of cutting at one width, so the mobile preview shows the text in a phone-width card. Lead with the words people search for, keep the brand at the end, and the important part shows on both.

The checker lists every og: and twitter: tag it finds and draws the link card for Facebook and LinkedIn and for X from them, so you can see at a glance whether a shared link will have a title, a description and an image. It flags a page with no Open Graph tags, a missing og:title, og:description or og:image, an og:url that is not a full address and a twitter:card value that is not a card type.

It also fetches the share image once and reports whether it loads, whether it is really an image, and its width and height, read from the file's own header. 1200×630 pixels, about 1.91 to 1, displays well on every network. For each app's own tag order and image rules, with cards for seven apps and how to refresh each one, use the social share preview.

A robots meta tag or X-Robots-Tag header with noindex keeps the page out of search results whatever its title says, so the checker flags it. A missing viewport tag makes phones show the page zoomed out, a missing lang attribute makes screen readers and translation tools guess the language, and a missing charset can garble accented letters.

For sites in more than one language, hreflang tags list each version of the page. The checker lists them and flags three common mistakes: a code that is not a valid language code (en_US with an underscore, en-UK where the region code is GB, or a region such as gb written alone), a set that does not list the page itself, and a set with no x-default for visitors whose language is not listed.