What is the difference between a canonical tag and noindex?

They answer different questions.

Noindex says: do not show this page in search results. Use it for pages that should exist for visitors but not in search: thank-you pages, internal search results, and thin utility pages.

A canonical tag says: of these similar URLs, this is the one to treat as the main version. It is a hint, not a command. Use it when the same content lives at more than one URL: tracking parameters, sorted or filtered lists, printable versions, or a page syndicated on 2 sites of yours.

Pick the right one for the job:

• Filtered product list with a parameter in the URL: canonical to the main category page.
• Order confirmation page: noindex.
• 2 URLs serving the same page because of a trailing slash or capitalization: fix with a redirect, and add a self-referencing canonical.
• Staging site: keep it behind a login. Noindex is a backup, not the lock.

Mistakes that come up all the time:

1. Noindex and a canonical to a different URL on the same page. The signals conflict. Choose 1.
2. Blocking the page in robots.txt and expecting noindex to work. If crawlers cannot fetch the page, they cannot see the tag. To remove a page from the index, allow crawling and use noindex until it drops out.
3. Canonical pointing at a redirected or noindexed URL.
4. Canonicalizing everything to the home page.

Check what Google actually chose in Search Console's URL Inspection tool. It reports the canonical you declared and the one Google selected. If they differ, the 2 pages are probably too similar, or the signals around them disagree.