Is duplicate content an SEO penalty?

Google has stated repeatedly there is no duplicate-content penalty. What happens instead is splitting: when two pages target the same query with similar content, Google picks one to rank and the other gets buried. The 'penalty' is the wasted authority on the non-chosen page and the underperformance of the chosen one. Consolidation fixes this — combined authority means better ranking than split authority.

What counts as 'duplicate' versus 'similar'?

Google uses a similarity score, not a binary threshold. Two pages with 90%+ identical text are treated as duplicates. Two pages with 60-90% similarity are flagged as cannibalising. Two pages on the same topic with different angles and 30-60% similarity are typically fine. The Site Crawler reports similarity scores so you can act on the right threshold.

Should I 301 or canonicalise duplicates?

301 when only one URL should exist going forward — the losing URL is removed from the index, link equity flows to the winner. Canonical when both URLs must remain accessible (e.g., parameter URLs for filtering, different language versions, tracking variants) but only one should be indexed. Canonicals are a hint to Google; 301s are an instruction.

My CMS auto-creates duplicate URLs — how do I stop it?

Common offenders: WordPress tag and category archives, Shopify collections that overlap with smart collections, Magento layered navigation. Fix at the source: in WordPress disable archives you don't need, in Shopify mark redundant collections as noindex, in Magento configure parameter handling in Search Console plus canonical rules in the catalogue config. Stopping the source beats cleaning up after.

How to Fix Duplicate Content

Duplicate content splits ranking signals between URLs that should be one. Google chooses one to rank and buries the others. The fix depends on the duplicate type: exact duplicates from CMS quirks need canonical tags, accidental duplicates need 301s, intentional near-duplicates need merging or rewriting. This guide walks through the full triage and the platform-specific patterns that prevent recurrence.

1. Generate the duplicates report

Step 1

Run the Site Crawler with similarity scoring

Run a fresh Site Crawler with the option Detect near-duplicates enabled. The crawler scores every page pair for content similarity and groups pages above the threshold (default 80%) into clusters. Export the cluster CSV.

Step 2

Sort by similarity

Highest-similarity clusters (95-100%) are exact duplicates from CMS quirks — fix first. 80-95% clusters are near-duplicates that need editorial decisions. Below 80%, generally fine unless intent overlaps significantly.

2. Classify each cluster

Type A: Same URL, different parameters

Example: /products/widget, /products/widget?utm_source=email, /products/widget?sort=price. The content is identical; only tracking or sort params differ.

Fix: Self-referencing canonical on the primary URL, robots.txt or Search Console parameter handling for query strings that shouldn't be indexed.

Type B: Multiple URLs, same content

Example: /blog/post-title and /posts/post-title serving identical content because the CMS exposes the post under two routes.

Fix: 301 redirect from the secondary to the primary. Update internal links.

Type C: Near-duplicates competing for the same query

Example: /best-running-shoes-2025 and /top-running-shoes-uk with 70% similar content targeting overlapping search intent.

Fix: Editorial decision — merge into one comprehensive page, differentiate intent clearly, or rewrite the lower-performer with a different angle.

Type D: Variant pages (intentional duplicates)

Example: Multiple language versions, mobile/AMP variants, country-specific pages.

Fix: hreflang tags for language/country variants, AMP-specific canonicals, never 301 these.

3. Apply canonical tags

Step 1

Add self-referencing canonical to chosen primary

In the <head> of the primary URL:

<link rel="canonical" href="https://yourdomain.com/products/widget">

Self-referencing canonicals are best practice on every indexable page. They prevent accidental duplication from tracking parameters and similar URL variants.

Step 2

Add cross-referencing canonical on duplicates

If duplicate URLs must stay accessible (e.g. parameter URLs needed for filtering), add canonical on each duplicate pointing at the primary:

<!-- on /products/widget?utm_source=email -->
<link rel="canonical" href="https://yourdomain.com/products/widget">

⚠️ Canonical tags are hints, not directives. Google may ignore them if the suggested canonical isn't actually similar to the canonicalising page. For non-negotiable consolidation, use 301.

4. Apply 301 redirects

Step 1

Decide the winner per duplicate cluster

For each cluster, the winner is the URL with the most external backlinks (check Search Console Links report) or the most current Search Console clicks. If both are similar, choose the cleaner URL or the one matching your current URL conventions.

Step 2

301 the losers to the winner

Apply 301 via your platform — Redirection plugin (WordPress), URL Redirects (Shopify), nginx rewrite, Apache Redirect 301. See the redirect chains guide for syntax per platform.

Step 3

Update internal links

Internal links still pointing at the loser URL now go through a 301 hop. Update them to point directly at the winner. Use bulk find-and-replace techniques from the broken links guide.

5. Use noindex for pages that must exist but shouldn't rank

Some pages are duplicates by design — search-result pages, faceted-nav variants, tag archives, paginated lists. They need to exist for users but shouldn't compete in the index.

Step 1

Add noindex meta

In the <head> of these pages:

<meta name="robots" content="noindex, follow">

follow lets Google still discover links from these pages; noindex keeps them out of search results.

6. Platform-specific recurrence prevention

WordPress

Yoast / Rank Math: enable canonical auto-generation for posts and products
Disable tag archives and date archives if you don't actively use them
Set noindex on author pages unless authors are part of your E-E-A-T strategy

Shopify

Products auto-canonicalise to /products/{handle} regardless of which collection link the visitor came through
Avoid smart collections that overlap heavily with manual collections
For variants: use product-level canonical, not variant-level URLs in nav

Headless / custom CMS

Build canonical tag generation into your routing layer, not per-template
Strip tracking parameters server-side or via JavaScript to standardise URL display
Add CI checks: any new template must emit a canonical tag

7. Re-run the audit

After consolidating, run the Site Crawler again. Duplicate clusters should drop significantly. Remaining clusters are usually intentional variants (language, AMP) or pages still being merged editorially.

🕷 Re-run the Site Crawler

Verify duplicate count has dropped after consolidation.

Run Site Crawler →