og:url differs from the canonical address

Why it matters

The Open Graph protocol defines og:url as the canonical URL of the object, used as its permanent ID. When it differs from the page's own canonical, the page states two different main addresses.

What to do

Look at the finding: it shows both addresses as written and as resolved, and which parts differ.
Set og:url to exactly the canonical address, same scheme, host, path, trailing slash and query.
If your SEO plugin fills og:url separately, make it follow the canonical.

When to leave it as it is: if you deliberately share a different address, such as a campaign landing page, you can leave it.

What we read

We read the <head> of each crawled page as served, without running JavaScript, and compare its one og:url with its one <link rel="canonical">, both resolved to full addresses; the canonical sent in an HTTP header is not used. A noindex declaration alone does not exclude a captured page from this comparison.

A page is reported when the two differ in scheme, host, path or query; a trailing slash, www. and the order of parameters all count as differences. When og:url differs from an https:// canonical only by being http://, on an HTTPS source page with an internal HTTP og:url, and the lingering http:// references rule is also selected, that rule reports this case instead. Otherwise this rule reports the mismatch.

A page without either tag is not judged; one with two of either, an empty value or an address we cannot resolve is marked as not checked.

Example & evidence

Illustrative example — one page head, not a live audit.

https://bakery.example/cakes/ has in its head:

<link rel="canonical" href="https://bakery.example/cakes/">
<meta property="og:url" content="https://bakery.example/cakes">

The two differ by the trailing slash, so the page is reported with both addresses.

<meta property="og:url" content="https://bakery.example/cakes/">

A possible correction: make og:url match the canonical exactly.

Confirm & check again

Check for yourself: view the page source and compare the resolved og:url with the resolved href of the head rel="canonical"; relative values and a <base> tag can make the raw strings look different even when the addresses match.

Check again: Full audit — run a new audit of the site. The finding closes when the page is crawled again with its head read in full and both tags name the same address, or one of them is removed. Adding a second og:url or canonical marks the page as not checked and keeps the finding open.

Limits & sources

The check compares the two declarations; it does not request either address or see which one previews or Google actually use.

Confidence: High because both tags are read directly from the page head.

Sources

ogp.me, The Open Graph protocol — defines og:url as the canonical URL of the object, used as its permanent ID. Checked 28 September 2026.

Editorial review: 29 September 2026

Links and lists still point to http:// addresses — takes the internal HTTP-only scheme difference when that rule is selected.
A page names a different page as the main one — the canonical itself points to a page that does not name itself.

The share address differs from the canonical address