How Google picks one URL from a set of duplicates, and how to influence it.
Aditya Challa1 min read
A canonical URL is the version of a page that Google picks as the representative one when several URLs show the same or very similar content. You can suggest it with rel="canonical", redirects or your sitemap, but Google treats these as signals, not commands.
<link rel="canonical" href="https://www.example.com/page" /> in the page headrel="canonical" HTTP header (useful for PDFs)HTTP vs HTTPS, www vs non-www, tracking parameters, and printer or sorted versions of the same page.
Do not use robots.txt or the Removals tool for canonicalization, and do not point canonicals at pages that redirect or return errors.
Related: /blog/how-do-i-submit-a-sitemap-to-google
Related: /blog/how-to-remove-a-url-from-google-search
Related: /blog/what-does-crawled-currently-not-indexed-mean
Lab evidence
Facts in this draft were checked on 12 Oct 2026 against the primary sources listed in the body (official docs, release schedules or Google/YouTube help pages). No benchmarks or timings were run, so the post makes no performance or measurement claims; dates are quoted as published by the maintainers, and any estimate is labelled as such.
Explore the product
Site owners working on indexing, rankings and organic growth
Why Google treats some 200 OK pages as missing, and the two correct fixes.
Summary12 Oct 20261 min read
New Ready Client Sites angle: ShopperCove fetched all 15 live prebuiltwebs.com demos — client-rendered React SPAs with hash routes, built-in admin/enquiry/upload APIs, placeholder content JSON, no robots.txt/sitemap, soft 404s — and turned it into a pre-purchase SEO and handover checklist. Distinct from the Node.js packs and launch-checklist posts. One disclosed hop. No purchase.
Summary5 Oct 20263 min read
Insights now lives inside Search Console. Here is what each card shows and how to act on it.
Summary14 Oct 20263 min read