Page 1 using query string parameter vs direct link

[borishar]

Hey guys,

On my website I have the following structure for the urls

domain/companyname/

each of these links have pages

domain/companyname/?page=2
domain/companyname/?page=3
domain/companyname/?page=4

they all have proper next/prev rels.

My question is about page 1. Instead of having 2 links for the same page

domain/companyname/ and domain/companyname/?page=1

should I 301 the page 1 to domain/companyname/ ?

Wouldn't this hurt the next/prev rels, since on the second page prev rel points to page 1

Thank you!

[Sand]

I'd just canonical domain/companyname/?page=1 to domain/companyname/ since the content is exactly the same on both.

[lucy24]

should I 301 the page 1 to domain/companyname/ ?

Not if the "page1" link originated on your own site. Nobody likes links that point straight to a redirect. If you can't tweak your code so the element ?page1 is replaced by "" (no query), then stick with "canonical".

I think it's safe to assume that any time you have an URL with a query string, the major search engines will try for-the-hell-of-it requesting the same path minus query, just to see what comes up. So you'd want a "canonical" anyway.

[borishar]

Thanks guys! Wouldn't rel canonical interfere with the next prev rels, which are on the same page pointing to the second page?

[Sand]

In a perfect world, the prev link that points to page 1 would be domain/companyname/ and not domain/companyname/?page=1. So if you can fix that, that would be best.

But if that isn't something that you're able to do, then implementing the canonical on page=1 is probably the best way to go.

[borishar]

Hey Sand,

I can do that, but what should I do with the page1 link. It pulls up the same results as /companyname/ and might be considered as duplicate page?

[lucy24]

what should I do with the page1 link. It pulls up the same results as /companyname/ and might be considered as duplicate page

That's why you say "canonical". It means "Yes, I know, I can't do anything about it so just index this form". It will take a little while for the existing search-engine index to update itself.

[ZydoSEO]

Ideally, you would:

1) fix all links on your own site so that they reference domain/companyname/ (without ?page=1).

2) 301 redirect domain/companyname/?page=1 to domain/companyname/ in case other external sites are linking to this version of the URL

3) implement the following <link rel=...> elements:

For page 1:

<link rel="canonical" href="domain/companyname/">
<link rel="next" href="domain/companyname/?page=2">

For page 2:

<link rel=canonical" href="domain/companyname/?page=2">
<link rel="prev" href="domain/companyname/">
<link rel="next" href="domain/companyname/?page=3">

Many would simply use a <link rel="canonical"> element to fix such issues because it's the "easy", lazy man's solution. But the canonical link element was really meant to be used as a last resort when you cannot fix canonicalization issues using 301 redirects like is often the case on ecomm sites that heavily depend on querystring parameters like ?catid=123&subcat=456&productid=789 and 301 redirects to deal with every ordering of those 3 querystring parameters would be very complex.

If you fix it with <link rel="canonical"> but have visible links to the non-canonical form of the URL (domain/companyname/?page=1) on the site and/or you have external links to the non-canonical URL, users will continue to link to the non-canonical (?page=1) version of the URL going forward since most users create links by copying what they see in their browser address bar. If they click on a non-canonical link on another site, their browser will STILL show the non-canonical so a copy/paste to create a link will continue to yield non-canonical links.

However, if you correct this with a 301 redirect, users will no longer see the non-canonical (?page=1) form of the URL in their browser even if a user follows a non-canonical link (internal or external) to domain/companyname/?page=1 because the 301 redirect will immediately change their address bar to domain/companyname/. Going forward, the URL users copy out of their browser address bar to create links will ALWAYS be the canonical form. So 301 redirects almost guarantee that all links going forward will be to the non-canonical URL.

[borishar]

Hi Zydo,

I will try implementing the 301. My question now is about the following statement:

<link rel=canonical" href="domain/companyname/?page=2">

Why do I need to canonical the page two to itself?

[lucy24]

Why do I need to canonical the page two to itself?

You don't ... unless there is a possibility that the "page 2" content could be viewed under some other URL, such as

?page=2&sortby=date&itemcount=30

If there is only one possible URL for a given page*, you don't need a "canonical".


* Where "page" = "content" or "what a human sees", as distinct from URL which is "what a search engine indexes".

[ZydoSEO]

The canonical link element is used independently of the prev/next link elements. As Lucy24 said, if your site uses (or there is a good chance that your site might use) other querystring parameters on the paginated pages for things like internal or external tracking codes, sorting, filtering, etc. then the canonical link element would help the engines by indicating the proper canonical form of each URL.