Duplicate page error from SEOmoz

[scanlin]

SEOmoz's Crawl Diagnostics is complaining about a duplicate page error. I'm trying to use a rel=canonical but maybe I'm not doing it right.

This page is the original, definitive version of the content:

https://www.borntosell.com/covered-call-newsletter/sent-2011-10-01

This page is an alias that points to it (each month the alias is changed to point to the then current issue):

https://www.borntosell.com/covered-call-newsletter/latest-issue

The alias page above contains this tag (which is also updated each month when a new issue comes out) in the section:

Is that not correct? Is the https (vs http) messing something up?

Thanks!

[RyanKent]

Your example of how the canonical tag is presented is perfect. You can verify your canonical tag is recognized using the MozBar Analyze Page > Page Attributes.

The https should not cause any issue.

Is it possible something else is causing the duplicate content warning? For example, I noticed when I enter the same URL with a trailing slash, the same page appears:

https://www.borntosell.com/covered-call-newsletter/latest-issue

https://www.borntosell.com/covered-call-newsletter/latest-issue/

Of course, both pages are canonicalized properly but it does show there is an issue on your site which should be addressed. Technically adding a slash to the end of a URL is a different page in much the same way that adding or removing the "www" subdomain offers a different page.

[scanlin]

Thanks for the response. A followup question.... when you say:

"when I enter the same URL with a trailing slash, the same page appears"

What would you expect to happen?

Should I 301 redirect the url with the slash to the url without the slash, or something else?

Thanks.

[RyanKent]

What would you expect to happen?

Well, when look at the current Q&A URL:

http://www.seomoz.org/q/duplicate-page-error-from-seomoz

now try adding a trailing slash to it

http://www.seomoz.org/q/duplicate-page-error-from-seomoz/

in this case the URL with the trailing slash is rejected, which is expected. What I don't expect is for the page to be redirected to the root Q&A page. Not sure why that was implemented here and it seems unusual to me.

An example of a site with the (close to) expected behavior is the Distilled site:

http://www.distilled.net/blog/seo/7-technical-seo-wins-for-web-developers/

If you try to remove the trailing slash, you will notice the URL is changed back to a URL with the trailing slash.

What I am a bit surprised about in the Distilled example is the redirected page shows a 200 header code. I believe it should display a 301 redirect from the page without the trailing slash to the page with it.

Overall I would say this is very nit-picky SEO but since the topic came up, it is relevant. Actually I am curious enough about this I will use my staff question this month on this topic

Should I 301 redirect the url with the slash to the url without the slash, or something else?

Yes. This issue is very similar to www vs non-www. Each page should have one version of a URL available and not more. It is not so important which you choose in as much as you make a decision and remain consistent.

The Distilled site chooses to end all URLs with a slash. That is a fine choice. SEOmoz chooses (as do I) to end category pages with a slash and actual web pages without a slash. Using this system if you look at a URL and it ends with a slash, you know you can go deeper. If the URL does not have a slash, you have reached the bottom.

Wow. This is the most time I have spent thinking about this little detail of SEO

[scanlin]

Thanks, Ryan.

My entire site is https. I 301 any http request to the https version. That works fine.

I also 301 any non-www to the www version. That works.

I'm not currently doing anything special with the trailing slash (or lack of). I guess I should 301 anything with a trailing slash to the non-slash version, rather than serve up slash and non-slash each as status 200.

Appreciate the guidance...