Overlapping content - is it duplicate ? and how to manage it

[Whitey]

Google Quality Guidelines
Does the site have duplicate, overlapping, or redundant articles on the same or similar topics with slightly different keyword variations?
[googlewebmastercentral.blogspot.com...]
and

Consolidating approximate duplication (if lots of pages on the site about the same topic)
[searchengineland.com...] Vanessa Fox

In the past Google just filtered same content for parent/child pages and displayed the best result. But things may have changed slightly. with Panda.

In the context of Panda , what is "overlapping content" defined as, and how should it be best managed? How for example are shopping sites with product dealing with this.

[deadsea]

Given the focus on content farms, I take overlapping content to be articles like:

Red Widgets
Best Red Widgets
Red Widgets are Best
Widgets in Red
Rouge Widgets
Where to get Red Widgets
What do do if you have a Red Widget
etc.

Content farms were literally writing a separate article about each and every variation on the search phrase, none of which were very deep.

[deadsea]

This is one of my pet peeves about Panda, btw. There should be a way to target your site to all those variations in search terms. As a webmaster you should be able to tell Google that these variations are so similar that my site has relevant content for them. Before panda, you would have to create pages targeting them. Post panda, Google tells us "creating overlapping pages is a sign of spam", but gives no guidance on how to structure a site to rank appropriately for all the variants. And let me tell you, even after Panda, Google doesn't rank a single high quality page in the top for all the possible search phrase variations that are relevant.

This is a case in which Google should set up a system to view webmasters as partners to creating better SERPs but it seems that Google almost always views webmasters as opponents, even when Google could benefit from listening to them.

[Whitey]

but gives no guidance on how to structure a site to rank appropriately for all the variants

Exactly right

I'd like to ramp this up for debate, as I think this is important and not been dug into deeply enough on other threads. There appears to be little explanation, or conversations around best practice, but that's understood because duplicate content alone was/is a potentially complex issue, let alone with Panda, so here goes :

I see 2 shopping sites with parent/child arrangements like this :

Site 1 - has content snippets repeated

www.domain.com/location-widget-list
30 word snippets on each widget, linking to child URL's

www.domain.com/sub-location-widget-list
30 word snippets on each widget, linking to child URL's

www.domain.com/child-of-sub-location-widget-list
30 word snippets on each widget, linking to child URL's
www.domain.com/widgets

www.domain.com/widgets/
Full 300 word description of widget, including 30 words used in snippet

Site 2

www.domain.com/location-widget-list
No 30 word snippets on each widget, but linking to child URL's

www.domain.com/sub-location-widget-list
No 30 word snippets on each widget, but linking to child URL's

www.domain.com/widgets/
Full 300 word description of widget


In the context of Panda which one is better?
If neither is good, what's the alternative?

[aristotle]

And let me tell you, even after Panda, Google doesn't rank a single high quality page in the top for all the possible search phrase variations that are relevant.

deadsea -- I doubt that Google does this intentionally. More likely it's a by-product of their effort to prevent content farm abuse.

[tedster]

It's hard to give you much more specific input without getting into an actual site review, which we can't do here. As I see it, the guidelines are really about what deadsea described in the first reply - it's bogus to create a separate page for each variation in the keywords. But it's not about accidental overlaps because of product categories and descriptions.

It's all in the intention. And it is most definitely shades of gray. I know of unpenalized sites that definitely chase one keyword variation per page. They use almost a classic doorway page scenario but, with calls-to-action (conversion points) right on the page instead of on the other end of a link.

[Whitey]

calls-to-action (conversion points) right on the page instead of on the other end of a link.

I couldn't visualise this - could you elaborate or describe an example.

[edited by: Whitey at 7:26 am (utc) on Oct 26, 2011]

[universetoday]

We built a tool that analyzes a site to identify potential overlapping content.

What we did was compare a collection of keywords against every page in the website. Each keyword will get a search ranking score for a page. So then, if two pages have a similar value for each keyword, then they're flagged as potentially overlapping.

For example, let's say you've got two pages:
site.com/widgets-blue/
site.com/blue-widgets/

If you searched your entire site for the keyword "blue widgets", those two pages would have similar relevance scores, and so, they'll be potentially overlapping.

You still need to use your judgement, though. Then you can decide if you want to delete the extra articles, or just ignore what the algorithm is telling you.

I'd be happy to let you use it if you wanted.

[deadsea]

calls-to-action (conversion points) right on the page instead of on the other end of a link.

I couldn't visualise this - could you elaborate or describe an example.

"click here to go to the sign up page" vs "Enter your desired user name here to sign up"

"click here to buy this item at amazon" vs "Add this item to your shopping cart"