Should every page have a canonical tag?

Yes. Every page should have one self-referential, absolute https canonical, unless it is a genuine duplicate that should point elsewhere.

What is a common canonical mistake?

Pointing the canonical at a redirect, a 404, a noindex page, or having every page canonicalize to the homepage. All of these make Google ignore it.

Document does not have a valid rel=canonical: how to fix it

Quick answer: This audit fails when the canonical link is missing, relative, or points to a redirect. Add one self-referential, absolute https rel=canonical per page that returns a 200 status. It consolidates duplicate-URL ranking signals so Google indexes and ranks the URL you actually want.

Lighthouse fails this audit when your page is missing a <link rel="canonical">, or has one that is malformed, relative, points to a redirect, or conflicts with other signals. The canonical tag tells Google which URL is the authoritative version of a page when several URLs serve similar content.

TL;DR

What: No valid <link rel="canonical"> in <head>, or it points somewhere Google can't trust.
Why it matters: Without it, duplicate URLs (tracking params, trailing slashes, http vs https, www vs apex) split ranking signals and Google may index the wrong one.
Fix: Add one self-referential, absolute, https canonical per page that returns 200.

What does the rel=canonical audit check?

Lighthouse verifies that the page has at most one rel=canonical link, that it is a valid absolute URL, and that it is not obviously broken (empty, relative-only, or pointing at a non-indexable target). It does not judge whether the chosen canonical is strategically correct. That part is on you.

Why does the canonical tag matter for SEO?

Most sites serve the same content under several URLs without realizing it:

https://example.com/page and https://example.com/page/
?utm_source=... and other tracking parameters
http://, https://, www., and apex variants
Paginated or filtered views of the same list

Google treats each as a separate URL. Ranking signals (links, engagement) get split across them, and Google may index a parameter-laden variant instead of your clean URL. A correct canonical consolidates all of that into one page.

How do I add a valid canonical tag?

Add a self-referential canonical to <head>:

<head>
  <link rel="canonical" href="https://example.com/audits/canonical-tag" />
</head>

Rules that make a canonical valid and effective:

Absolute URL. Use the full https:// URL, not /audits/canonical-tag.
Self-referential by default. Each page points at its own preferred URL. Only point elsewhere when the page is a genuine duplicate.
Resolve to 200. The canonical target must be indexable: no redirect, no 404, not blocked by robots.
One per page. Multiple conflicting canonicals make Google ignore all of them.
Consistent with other signals. The canonical, the sitemap URL, internal links, and any hreflang entries should all agree.

What are common canonical mistakes?

Canonical points to a redirect. If /page canonicalizes to /page/ and /page/ 301s back, Google sees a loop and ignores it.
Relative URLs. Some frameworks emit href="/page". Always render the absolute URL.
Every page canonicalizing to the homepage. A frequent CMS misconfiguration. It tells Google your inner pages don't exist.
Canonical to a noindex page. Contradictory: you're saying "rank this" and "don't index this."
Mismatched protocol or host. Canonical says http://www. while the live page is https:// apex.
Parameter URLs canonicalizing to themselves. ?utm_source=twitter should canonicalize to the clean URL, not to the tracked one.

How do I set the canonical in Next.js, WordPress, or Astro?

Next.js (App Router)

export const metadata = {
  alternates: { canonical: 'https://example.com/audits/canonical-tag' },
};

Next.js (Pages Router)

import Head from 'next/head';

<Head>
  <link rel="canonical" href={`https://example.com${router.asPath.split('?')[0]}`} />
</Head>

WordPress

Yoast SEO and Rank Math emit a self-referential canonical automatically. Do not also hardcode one in the theme: two tags conflict. Override per-page only for genuine duplicates.

Astro

---
const canonical = new URL(Astro.url.pathname, Astro.site).href;
---
<link rel="canonical" href={canonical} />

How do I verify the canonical is valid?

Re-run Lighthouse: the audit should pass.
View source (Cmd+U) and confirm exactly one rel="canonical" with an absolute https URL.
Open the canonical URL directly: it must return 200, not redirect.
Google Search Console → URL Inspection → check "User-declared canonical" matches "Google-selected canonical".

Related audits

Document does not have a meta description, the search snippet
Document does not have a title element, the primary SEO signal
Does not use HTTPS, protocol consistency feeds canonical correctness

Audit your URL at https://lighthouse-md.com.

Audit your page now

Paste your URL, get scores plus a CLAUDE.md plan for Claude Code.

Run audit →