Reference guide · technical-seo · Published 2026-08-16 · 4 min read
Canonical vs redirect vs noindex
Compare canonical tags, redirects, and noindex for duplicate content. Learn why a 301 beats a canonical, when they conflict, and the audit order.
- ·Three signals
- ·Which wins
- ·The practical table
The three signals and their strength
Duplicate and near-duplicate pages get handled with three different signals, each with a different position in the browser and the search engine:
| Signal | What it says | How strong |
|---|---|---|
| 301/308 redirect | 'This URL is gone; the content moved here' | Very strong; the browser follows it |
rel="canonical" | 'This URL exists, but this is the version to index' | A hint, only honored when Google agrees |
meta robots="noindex" | 'Do not index this URL' | Strong for where it is used, weaker for consolidation |
A redirect is the strongest because it acts on every request and every crawler. A canonical is a statement, and Google explicitly says a canonical is a hint that can be strengthened, weakened, or overridden. noindex is the backstop for a URL you do not want indexed but do not want to 404.
Which signal wins when they conflict
- Redirect wins over canonical: 301, followed by 302 and 308, prevent the page from being canonical'd. The canonical rule for this is that a redirect changes what the browser and crawler see; a canonical only describes the current page.
- noindex beats canonical: if the canonical page carries
noindex, Google does not index it regardless of the canonical it declares. The canonical merely names a partner;noindexis the instruction the page itself gives. - canonical pointing at a redirecting URL: malformed and merged against a target that moves. Best practice is that canonical and redirects point at the same final URL.
The decision table
| Situation | Use | Why |
|---|---|---|
| Old URL that still has visitors, bookmarks, links | 301 redirect | Consolidates equity and needs no canonical for the old URL |
| Same content reachable at two URLs (parameters, www, http) | 301 or canonical | A redirect resolves it for everyone; a canonical is the fallback for pages that must stay reachable |
| A page that should never appear but should stay reachable | noindex | Blocks indexation without a 404 or redirect |
| Duplicate content that must keep its own URL for tracking or click tracking | canonical pointing at the master | Wholly-owned duplicate, canonical is the only tool |
The canonical as the weaker hedge, in the real world
Google merges duplicate signals in a direction it decides per query. A canonical tells it the preferred page, but it can still:
- Pick a different URL from the set it sees (which happens when the canonical target has issues or little authority and the duplicate has links).
- Ignore the canonical on syndicated content it believes is sourced from elsewhere.
- Reindex slow-moving pages even with a correct canonical.
The consequence is practical: use a canonical when a redirect would break the user flow (tracking URLs, catalog filters, faceted navigation), and prefer a redirect whenever the URL is truly obsolete. A site that 301s dead pages and canonicals living pages covers both classes of duplicate, which is the canonical guide and the 301 redirect article handle from their own angles.
When noindex is part of the answer
noindex is not a substitute for consolidation, but it is the correct tool for a page that should stay reachable without competing: a filter URL, a print view, a thank-you page, or a thin variant you deliberately keep. The mix:
noindexto stop indexation without a 404,canonicalwhere the page is worth keeping as the indexed favorite,301when the page is a leftover that deserves to die.
The three are orthogonal but complementary, and the noindex guide and the duplicate content article spell out the cases where noindex is the right final signal.