During a Google SEO office hour gathering, Google’s John Mueller was asked whether the rel canonical tag or the noindex tag is the best way to deal with duplicate and streamlined content in e-commerce sites. John Mueller discussed these two options, and then proposed a third approach.
No index instruction
The noindex meta tag is a directive, which means that Google must comply with the meta tag and remove the page from appearing in search results.
What the noindex tag does is make the page no longer appear in Google’s search results.
Google’s official document state:
“You can prevent pages or other resources from appearing in Google Search by including the noindex meta tag or header in the HTTP response. When Googlebot next crawls the page and sees the tag or title, regardless of whether other websites link to the Page, Googlebot will completely remove the page from Google search results.”
Relative norms
The rel=canonical tag is a hint, not a directive. It provides Google with suggestions for which URL you want to display in search results.
advertise
Keep reading below
This is useful when there are multiple similar pages, especially when a shopping CMS generates multiple pages for the same product, usually the only difference is small things such as product color.
Google’s official rel canonical document explains the problem like this:
“The canonical URL is the URL that Google considers to be the most representative page among a set of duplicate pages on your website. For example, if you have the URL of the same page (example.com?dress=1234 and example.com/dresses/1234) ), Google will choose one of them as the specification.”
The rel specification is a useful solution because it can merge all links and relevance signals back to the homepage that the publisher wants to show in search results.
However, because Google treats the rel canonical tag as a reminder, there is no guarantee that Google will comply with it, and Google algorithms may decide to display other pages in search results.
advertise
Keep reading below
Rel Canonical and Noindex
The person who asked the question wanted to clarify whether it is best to use noindex or normalization.
Confusion is not an unreasonable thing, because any solution can be used to create a case.
This is the problem:
“We have a website… an e-commerce store with many product variants, and the content of these variants is sometimes repetitive.
So… I list all URLs that we want to keep or that we want to index… Then I list all URLs that we don’t want to be indexed.
The more I study, the more I will ask myself this question, normalization or no index?
I don’t know those will be better. “
Mueller replied:
“…I think the general question of whether I should use noindex or rel canonical for another page may not have an absolute answer.
So this is a bit random. Like if you are struggling, you are not the only one you like, oh, which one should I use?
This usually means that both options are fine.
So usually what I see is your really strong preference.
If the strong preference is that you really don’t want to show this content in the search, then I would use noindex.
If your preference is that I really want to combine all the content in one page, and if a single page appears, for example any content, but most should be combined, then I will use the rel specification.
The final effect is similar, well, it is likely that the page you are viewing will not be displayed in the search.
But for noindex, it will never be displayed.
For the rel specification, it is more likely not to be displayed. “
The third way to deal with duplicates and thin pages
Mueller next suggested that publishers can use noindex and rel canonical at the same time in order to benefit from both.
Mueller said:
“…You can do both.
One more thing… For example, if external links point to this page, putting them all there helps us figure out that you don’t want this page to be indexed, but you also specified another page.
So maybe we can move some signals forward. “
The combination of Rel Canonical and Noindex is not a frequently discussed solution. But according to John Mueller, this is an effective way to deal with repetitive and streamlined content.
But in the end, it’s up to the publisher to decide whether it’s important to integrate links and relevance signals, and whether it’s most important to ensure that the page does not appear in the search based on the results they want.
Citation
Google’s Noindex official document
Use noindex for block search index
advertise
Keep reading below
Google official document Rel Canonical
Which is best: NoIndex or Rel Canonical?
Watch the minute mark at 16:49
