Last updated:
Meta Tag Checker
This meta tag checker reads a page's title, description, Open Graph, X card, canonical, robots and hreflang tags.
How to use the meta tag checker
Enter a page's address and choose Check. Our server fetches the page as SchemaBizBot, sends its HTML back, and the checker reads its tags here in your browser. Or choose "Paste HTML" and paste the page's source, or just its <head>; pasted HTML is checked without leaving your browser. "Try an example page" loads a blog post's head with a handful of mistakes to find.
A URL check does a little more than read the page. If the canonical link names another page, the checker fetches that page too, to see that it loads and can be indexed. It reads the first 64 KB of the page's og:image and twitter:image to learn their format and real size in pixels, without keeping or showing the images. And it can fetch the page's hreflang alternates to see whether each one links back. A link such as /seo/meta-tag-checker/?url=https://example.com/ runs the check when it's opened, and "Copy a link to this check" makes one.
What each tab checks
The results open on the Overview. Each tab's badge counts its errors and warnings, and every finding is labeled Error, Warning, Note or OK, so the tabs say what's right as well as what's wrong.
- Overview: a summary of every area, whether the title fits a desktop result and the description a phone result, and two generic link previews built from the tags. For widths in pixels and where each would be cut, use the title tag checker and the meta description length checker.
- Open Graph: the four required tags, absolute addresses,
og:urlagainst the canonical, and the image's real size and format. The Open Graph checker opens here. - X card: the card type, the tags X reads and the Open Graph values it falls back to, with X's length and image limits. The Twitter card validator opens here.
- Canonical & robots: the canonical link and
Linkheader, where they point and whether that page works, plus the robots meta tags andX-Robots-Tag. The canonical checker opens here. - Hreflang: every language code, the self-reference, x-default, and a matrix of which version links to which. The hreflang checker opens here and also takes a sitemap.
- All tags: every
<title>,<meta>,<link>and<base>in the order the page has them, with the ones that ended up outside<head>marked.
Meta tags Google reads, and the ones it doesn't
A page's <head> serves two kinds of reader. Search engines read a few tags to decide whether and how to show the page; social sites and chat apps read others to draw a preview when someone shares a link. The checker covers both, and keeps them apart.
For Google, the tags that matter are short. The <title> is the usual source of the title link. The meta description is sometimes used for the snippet. <meta name="robots">, or <meta name="googlebot"> for Google alone, can keep a page out of the index (noindex, or none, which also means nofollow) or limit its snippet. The canonical link names the version of a page to index, and hreflang links name its language versions. Google also reads the viewport, the character set and a meta refresh, though for a redirect it recommends a server-side 301 instead.
Some tags are still common but do nothing in Google Search. The keywords meta tag has no effect on indexing or ranking, so the checker notes it rather than warning. The lang attribute on <html> doesn't tell Google the page's language either: Google works that out from the text. The checker still notes a missing lang, because screen readers and translation tools use it.
Tags for link previews
Open Graph tags (og:title, og:image and the rest) are what Facebook, and many other sites and apps, read when a link is shared. X reads its own twitter: tags first and falls back to Open Graph for the title, description and image, so a page with good Open Graph tags needs only a twitter:card line to choose the card's layout. The checker's previews follow the same order: Open Graph first, then the page's title and description where a tag is missing.
Common meta tag mistakes
- A relative
og:image./images/share.jpgworks in the page but not in a preview: the sites that read it don't resolve it against your address. The checker reports it as an error. - An image that's too small, or a placeholder. Facebook's minimum is 200 × 200 pixels and it recommends 1200 × 630; a 1 × 1 tracking pixel left in a template gives every share a blank. The checker reads the real size, not the size the tags claim.
- Two canonical links. A theme and a plugin each add one, naming different addresses. Give the page one.
- A canonical that points at a page that redirects, is gone or is noindex. The checker fetches the target on a URL check and says which.
noindexleft on after launch. A staging setting copied to the live site keeps the page out of Google. The Canonical & robots tab reports it from the meta tag and theX-Robots-Tagheader.- Hreflang without a self-reference, or with made-up codes. Every version must list itself, and
en-UKisn't a region Google reads: the United Kingdom isGB.
What the example page shows
"Try an example page" loads the head of a pour-over coffee guide, written with some of those mistakes. Its title, 53 characters, fits a desktop result, but its description, 116 characters, is too wide for a phone. Its og:image is a relative address, so the Open Graph tab shows an error, and og:url is missing. There's no twitter:card, so the X card tab explains that X may show a summary card from the Open Graph tags. On the Hreflang tab, the page lists German and en-UK versions but not itself, and has no x-default. Because it's pasted HTML, the image isn't fetched: a URL check is what reads an image's real size.
Meta tags and structured data
Meta tags describe the page; structured data describes the things on it, such as a product, an article or a recipe, and can earn rich results in Google. They're checked separately: the Schema Markup Validator reads a page's JSON-LD, Microdata and RDFa, and the SERP simulator shows how the title, description and rich results could look together in a search result.
Frequently asked questions
FAQ
Which meta tags does Google use?
For search results, mainly the <title>, the meta description (sometimes, for the snippet), the robots and googlebot tags, the canonical link and hreflang links. Google also reads the viewport and charset tags and a meta refresh. It doesn't use the keywords meta tag at all, and it detects a page's language from its text rather than from the lang attribute. Open Graph and X card tags are mainly for link previews, although Google can use og:title for a title link and og:site_name for a site's name.
What's the difference between the meta tag checker and the other checkers here?
This page checks everything at once and opens on an overview. The Open Graph checker, Twitter card validator, canonical checker and hreflang checker are the same tool opened on their own tab, each with a guide to that topic. Title and description length have their own checkers, with the pixel rulers.
Why can't the checker see my images or alternate pages when I paste HTML?
Pasted HTML stays in your browser: nothing in it is sent anywhere, so the checker doesn't fetch the images, canonical target or alternate pages it names. Check the page by its URL instead, and our server fetches those too, as SchemaBizBot.
Does the meta tag checker store my pages?
No. Pasted HTML never leaves your browser, and a fetched page isn't kept. Like every tool here, the checker records what kind of check ran, such as how many required Open Graph tags are missing, the twitter:card type and the codes of the problems found, never a tag's text. For a URL check it also records the checked address without its query string, as the privacy policy describes.