Taking this approach means that all pages can still free turkey number for whatsapp exist on the site, but all but one page will be de-indexed, thus solving cannibalization issues.
That said, canonicalization should be preferred over noindex, since ranking signals are attributed to the canonical, whereas otherwise they would not be.
Use this approach with caution.
Don't make the mistake of combining canonicalization and noindex, as the two signals are contradictory information, as John Mueller stated:
Noindex (only) and disallow in robots.txt (in general) are not clear signs of canonicalization. Simply having a noindex on a page does not tell us that you want to combine it with something else and that signals should be forwarded. You should not mix noindex and rel=canonical - they are contradictory information to us. We will generally choose rel=canonical and use that over noindex, but any time you rely on computer script interpretation, you reduce the weight of your input
(and SEO is largely about telling computer scripts your preferences.)
If you’re not careful, you’ll end up sending conflicting signals. Noindex can be useful when you need to fix cannibalization issues caused by thin content and duplicate content that has no inbound links and receives no organic traffic; think of your blog tag pages, for example.
Re-optimization of pages
You may find that you have inadvertently caused metadata cannibalization by not optimizing for keyword variations.
For example, let's say you run an eCommerce store and you sell a product in three different colors. It's common to see variations of the product on unique URLs but with the same title tag, H1 tag, and no clear differentiation between the variations except the image.
In this case, you can reoptimize your pages to indicate clear variations and address cannibalization issues.