One article can be reachable through more than one URL: a clean address, a tracking-parameter version, or an older address that redirects. A canonical URL is the version a publisher wants search engines to treat as the preferred representative of duplicate or very similar pages. Reviewing those signals helps you find conflicting instructions, but it does not guarantee indexing or a particular ranking.
This guide is for independent publishers maintaining an existing site. It includes a review worksheet, a hypothetical example and a practical check on one Congrid article. For the broader distribution workflow, see Build a Content Distribution Plan for an Independent Website.
Choose a preferred address first
For each article, record the exact HTTPS URL you want readers and search engines to use. Consider the hostname, path, trailing-slash policy and whether a parameter changes the actual content. A tracking parameter often creates another way to reach the same article; a genuinely different translated or regional page needs its own careful treatment. Do not assume every parameter is disposable.
Google's canonicalization guidance explains that Google selects a canonical using multiple signals. Your declared canonical is a preference, not a command. Google may choose a different URL, and a canonical declaration does not itself cause an article to be indexed.
Compare the signals
| Signal | What to inspect | What to record |
|---|---|---|
| Page response | Does the preferred address load the intended article? Do old variants redirect where expected? | Observed URL and response or redirect destination. |
| Canonical declaration | Does the HTML head contain one valid canonical link to the intended absolute URL? | The declared target for each inspected version. |
| Internal links | Do article cards, navigation and relevant related articles use the preferred address? | The page linking to an inconsistent variant and its destination. |
| Sitemap | Does the sitemap list the preferred article URL? | Exact sitemap entry and the date checked. |
| Search Console | What does URL Inspection report for the inspected URL and Google's chosen canonical, when available? | Inspection time, reported state and any unavailable fields. |
Google describes redirects and canonical link annotations as stronger signals, with sitemap inclusion a weaker signal. These signals work best when consistent. A sitemap submission is a hint to Google and does not promise that a URL will be crawled or indexed. See Google's canonical selection guidance and sitemap guidance.
A canonical link for an HTML article can look like this, with your own preferred URL substituted:
<link rel="canonical" href="https://example.com/blog/example-article">
Place it in the HTML head. This example is an illustration, not a change to your website. If a CMS already generates canonical tags, check its output before adding another tag; conflicting declarations make diagnosis harder. Google recommends absolute URLs for this annotation.
Worked example: a campaign link and a clean article URL
Imagine a publisher shares an article at https://example.com/blog/garden-notes?utm_source=newsletter. The underlying article is also available at https://example.com/blog/garden-notes. The team chooses the clean URL as the preferred address after confirming that the parameter does not change the article.
The reviewer opens both addresses, records their rendered content and declared canonical, then checks the article listing and sitemap for the clean URL. If the parameter version also declares the clean URL and the internal links and sitemap agree, the signals are aligned. If the parameter selects genuinely different content, do not automatically consolidate it: investigate the purpose and consult the platform's documentation.
This is a hypothetical review, not a claim about a real site's traffic or Google selection. Capture the actual URLs and observations in your own change log; do not infer Google's selected canonical from the page's declaration alone.
Review one live Congrid example
On November 12, 2026, the Congrid drand core guide loaded in English and declared its clean URL as canonical. Its English language link led to the same path with the lang=en parameter; that version also declared the clean article URL as canonical. These are browser observations of publisher declarations, not a statement that Google has indexed the page or selected that URL. Recheck current behavior before relying on a past observation.
The Congrid blog provides article links, while the publisher guide explains website participation. Search-engine indexing and Congrid website verification are separate processes.
A repeatable publisher checklist
- Inventory the variants. For one important article, list its clean address, actual internal links, campaign links and any known old address. Record how you found each one.
- Compare content and response. Open each URL and note whether it loads the same article, redirects, errors, or serves meaningfully different content. Keep language and regional intent in view.
- Inspect the declarations. View the rendered page's canonical link and any applicable HTTP canonical header. Record the complete target rather than just the domain. Check for conflicting declarations.
- Align your own signals. Where your site controls them, use the intended URL consistently in internal links, canonical declarations and sitemap entries. For retired URLs, assess redirects with the person who manages deployment.
- Check Google separately. In Search Console, inspect the exact URL and record the reported indexing status and user-declared and Google-selected canonicals when the report provides them. A live test and an indexed report answer different questions.
- Recheck after changes. Confirm the public response and declarations after deployment, then allow time for recrawling. Record the date and owner of the follow-up.
Diagnose a mismatch
| Observation | Next question |
|---|---|
| The declared canonical points to an unrelated article | Is a template or CMS setting inserting the wrong target? |
| A variant redirects to one URL but declares another | Which destination is intended, and can the signals be aligned? |
| Sitemap and internal links disagree | Which version should be promoted, and which template or feed supplies the other link? |
| Google selects a different canonical | Are the pages actually duplicates, and what other signals or crawl evidence might explain the selection? |
| The URL is not indexed | Is it discoverable and indexable, and what reason does Search Console report? Do not assume a canonical tag is the cause. |
Avoid using robots.txt as a canonicalization mechanism: blocking crawling can keep Google from seeing page-level signals. Likewise, do not use noindex as a routine way to select the canonical within a duplicate set; follow Google's specific guidance for the situation.
Keep the review proportional. Start with one important URL and its observed variants, correct conflicting signals you control, and monitor the actual result. Measure organic search visits, referral visits and Congrid protocol outcomes separately; a tidy canonical setup is not evidence of any specific traffic gain.