Check every meta tagthe way Google reads it.
Paste a URL and get its title, description, canonical, robots, hreflang, Open Graph, X card and structured data audited in seconds, with a Google preview, a share card and the exact tags to fix it.
Enter a URL to audit its metadata
Title, description, canonical, robots, hreflang, Open Graph, X cards, structured data and more, each with a verdict and the fix.
What the checker reads
Everything a crawler takes from your page before it decides how to show it, grouped the way you would fix it.
Title and description
Measured in pixels, not just characters, because that is how Google truncates. You see the result exactly as it will be cut.
Indexing signals
Meta robots, the X-Robots-Tag header, robots.txt for this exact path, the HTTP status and every redirect. Any one of them can keep a page out of Google.
Canonical and hreflang
Which URL Google should index, and which language version goes to whom. Conflicts, relative URLs and invalid codes are named, not just flagged.
Open Graph and X cards
The tags behind every link preview, plus the real share image: its status, type, weight and pixel size, as a crawler downloads it.
Structured data
Every JSON-LD block is parsed. A block with a stray comma is invisible to Google, and you get the parser error to find it.
Why a crawler sees something your browser does not
Most "my tags are right there" arguments end with one of these. The checker reads the raw HTML the server returns, so it catches all four.
Tags added by JavaScript
Your inspector shows the page after scripts ran. Social scrapers never run them, and Google only does later and not always. If a tag is not in the HTML, it is missing.
A head closed too early
One invalid element in the head, an image or a div injected by a plugin, ends it on the spot. Every meta tag after that point lands in the body and is ignored.
A noindex nobody can see
The X-Robots-Tag header never appears in the source. A staging setting left on in production is enough to drop a whole site from Google.
A different page for bots
Firewalls and some CDNs answer bots differently. Fetch as Googlebot or as Chrome to compare what each one receives.
Titles are measured in pixels
Character limits are a rule of thumb. What Google actually cuts on is width, so a title full of W and M runs out long before one full of i and l.
Title: about 600 pixels
Roughly 50 to 60 characters in practice. Put the words that earn the click first, the brand last, so a cut costs you the least important part.
Description: about 160 characters
Google rewrites many descriptions, but it uses yours far more often when it matches the page and the search. Write it as a reason to click, not a list of keywords.
The mistakes we see most
Each of these is silent: the page looks fine in a browser and loses traffic anyway.
Canonical pointing at the homepage
A template that hardcodes the canonical tells Google every page is a duplicate of the home page. Only the home page gets indexed.
Relative og:image
A path like /images/share.jpg works in your browser and fails on Facebook, LinkedIn and WhatsApp. Share images need a full https:// URL.
Staging noindex in production
The single most expensive line of HTML. Check the header too: many hosts set it from an environment switch.
One title for the whole site
When every page has the same title, Google rewrites them all and none of them reads like an answer to the search.
Only care about how the link looks when shared? Open the Link Preview Checker
Sharing the link in a campaign? Tag it with the UTM Builder
Your metadata is right. Now track who clicks.
LinkScale turns the links you share into branded short links and landing pages, with the analytics to see which post, bio or campaign actually brings visitors.
Branded short links
Your own domain, a clean slug, and the same preview tags carried over so the card stays right.
Click analytics
Visitors, countries, devices and sources for every link, in real time, without adding a script to your site.
Opens in the right app
DeepLinks take mobile visitors out of the Instagram and TikTok browsers and into the app or the page that converts.
Metadata checker: questions
Still have questions?
OnlineReal humans, usually replying in a few minutes.
It is free, with no account. The URL is fetched by our server because a browser cannot read another site for you, and the result is sent straight back to your browser. Nothing is logged, written to a database or passed to a third party. The paste mode goes further: your markup is parsed in your browser and never sent at all.
Go further with LinkScale
Get the click, then measure it
Good metadata earns the click. LinkScale shows you where every click came from, with branded links and landing pages built for social traffic.
Free 14-day trial on every plan