SEO Tycoon

**littlemoomin** · 04-08-2011, 02:54 AM

What's the best way to overcome this internal dup. content problem?

We have what are essentially product type pages with a lot of content generated by AJAX, without its own URL.

But within each product page, users can also filter by different specs, which reduces some of the content, but still has a huge amount of the same content as the main level page.

An example:

An example main product page - Ford | Focus | Reviews | Carbuzz

One filtered by spec (engine type) - Ford | Focus | Reviews | Carbuzz

As you'll see the content that a spider would see is v.similar.

We can't redesign the site. Would it be a good idea to add rel=canonical to the engine level pages, so we have one decent page (the main car page) with a lot of content that doesn't appear on other of our URLs? Or is there another way to solve this?!

Thanks!

James

**angilina** · 04-08-2011, 07:49 AM

I think adding canonical link element is a good idea.

Also, see if parameter handling can help:

http://www.google.com/support/webmas...&answer=147959

http://www.mattcutts.com/blog/canonical-link-tag/

**chrishirst** · 04-08-2011, 10:41 AM

If they are

Quote:

generated by AJAX, without its own URL.

It's not even a consideration as "spiders" won't even know they exist.

**teegoWebs** · 04-08-2011, 03:09 PM

The only way google can find something is if it follows a link to it.

You can submit a sitemap to google. This will tell the bots to only search those pagse for information. You can google sitemap generator and just edit out what you dont want.

But yeah chrishirst is right, certain programming languages are not picked up by googles bots.

**smsinhindi** · 04-09-2011, 01:59 AM

rel=canonical seems better idea.. Google has posted on similar issue here -
http://googlewebmastercentral.blogsp...canonical.html

**vivekar** · 04-09-2011, 02:58 AM

Quote:

We have what are essentially product type pages with a lot of content generated by AJAX, without its own URL.

Canonical would help you point the dups to the original page.
If you still want google to index the additional content, you can try considering delivering the content via HashBang URIs.

Example: URL/category/#!filter

Be cautious though.

How To: Allow Google to Crawl your AJAX Content
Hash-bang: Good for the web but isn't ready for prime time
Gawker Learns the Hard Way Why ‘Hash-Bang’ URLs are Evil

Hope that helps you.

**littlemoomin** · 04-11-2011, 04:12 AM

Cheers guys!

We get about 2/3rds of visitors to engine specific pages. Do you think removing them from the index (using rel=canonical) would mean the parent pages (model pages) would in time replace those pages with similar search results?

**chrishirst** · 04-11-2011, 12:19 PM

No and why would you want to remove pages that bring search traffic?

**littlemoomin** · 04-13-2011, 11:37 AM

Because if they are causing the other pages to not rank well as it's repeat internal content

**chrishirst** · 04-13-2011, 03:05 PM

Ok so your idea is to STOP some pages ranking in the hope of getting other to "rank".
You do realise that it doesn't actually work like that don't you?

All that you are likely to accomplish is to remove your pages that are appearing in the SERPs, and allow somebody elses pages to move up.
Do these pages also get traffic? Or is it simply "rank checking" that has thrown up this "problem"

**suichiro** · 04-13-2011, 03:36 PM

If I'm thinking it right, we used to have the same problem on our site. There might have been something wrong in your site's coding that regards to the right call of page. Generate a sitemap.xml for your site, and check those URL that seem to be duplicated if they really should be existing.

**littlemoomin** · 04-17-2011, 05:45 AM

Could another way be to test it by blocking say 10 pages with robots.txt and seeing if that makes any difference?

**chrishirst** · 04-17-2011, 08:57 AM

Test it by all means, just bear in mind that if it does have an adverse effect, recovery may take a long while, if it ever does.

AND how exactly are you going to block a page that doesn't have a URL to block??????

**LarryHeart** · 04-18-2011, 10:06 AM

The pages which bring traffic must be the main and have canonical attribute.