Confusion About Duplicate Content

Ok, so my understanding of duplicate content is that when Google looks at a page that has 100% original content, it considers that content to be important, and once indexed, would consider that content to be the original.

If someone then decides to duplicate that content, Google will notice that it is not the original as that same content has already been indexed on another webpage, and then Google won’t consider that duplicate content to be important, so it would result in the duplicate content not being indexed or appearing deep in the search results.

But what happens if 50% of content is duplicate? I have noticed that the content is not considered duplicate, but would Google still prefer 100% unique?


The reason I’m asking this is because I want to put duplicate snippets of text on several of my internal pages, and I’m wondering how Google would see this.

I've long wondered this. How unique is unique and on what basis does Google decide if content is duplicate or not. This article seems to have a answer, but it is not the complete picture:
(cant find damn article, its about an SEO creating many many user profiles with sparse content and set with the task of ranking as many of them as possible, without running into duplication issues, ahhh well I'll leave you with a far better article I just found on my search, that probably explains it way better than this post: http://www.seobythesea.com/?p=999 (New Google Process for Detecting Near Duplicate Content 2008))

Answer: Google decides by certain characters (like "." a dot) what is a piece of content, and then compares that to the other content to see if it is duplicated).

On pure conjecture: Some duplicate sentences are normal or even beneficial - which newsarticles centered around a certain quote, doesn't list this quote in the article? Political newsarticles are rife with quotes. Now which blogpost is more relevant for a user, rest of the factors being equal: the one with or one without the text of a certain controversial tweet? The scientific article, with or without the citation of the expert? I think the duplicated content becomes semantically relevant to the search term. I think you can calculate that and rank those articles accordingly.

Google can take (part of) a sentence and compare it to its database. Perform linguistic analysis on the total corpus. Semantic analysis of html structure like on http://www.seobythesea.com/?p=3824 where they translate a structure like:
<h2>SEO Facts</h2>
<ul>
<li>Exciting</li>
<li>Stands for Search Engine Optimization</li>
<li>Is about optimizing sites for users and search engines</li>
</ul>

into logical structures. Now compare the next article that writes about SEO and mentions "search engines" and "Optimization" isn't more relevant for us, and for Google, than an article that doesn't mention those terms. In closing: The meaning is duplicated and rewarded, literal duplication of syntax is easily spotted and dealt with accordingly.

I think what is more important is to consider what else you can add to that content to make it stand out. For instance I have had much success with ecommmerce sites that sell basically the same products that many other sites do

With the feeds that are used we get the same generic descriptions, title, etc that are used across the board on many other sites. However, where we have seen the best return is when we add our own snippet of information about why we recommend the product or maybe an editorial review or further description. That additional content coupled with ugc (user reviews) gives us some very nice placements as well as a better user experience which tends to lead to more registered users, unique reviews, etc that give us the extra boost that is not present on the products that we use the stock, duplicate content.

I am not sure about a percentage or anything like that of what you would need to do, but we just follow the rule that if we can provide the best info for what we have on the page then we are doing ok, even if some of that is duplicate.

There are several ways to include your text snippets so that they are not indexed to avoid repeating them on multiple pages. That's what I'd do in this situation. Make your indexed text original and unique.

Back last summmer, I pointed to a great Infographic on just this topic here --
http://www.canuckseo.com/index.php/2010/07/seo-infographicselliance-rocks/

From what I know before and since that post, filled in by 2 conf and lotsa late night beers with other speakers etc...incl of course Rand's great comments on same over at seomoz...

Dup content filters STILL exists....and with the new Google BLEND, it appears that even reviews are suffering to a degree -- at least up here in g.ca land they are...

Who'd try to use a duplicate review, same exact english from the same exact poster for their diff car dealerships? Well, it appears at least a few canuck car dealers figured they'd not get caught!

NICE CATCH Google....

:-)

Jim

PS makes one wonder tho, what kind of a "sales pitch" you'd have to present to a multimillionare car dealership Principal/Owner that you can provide SEO serp success with duplicate reviews....sigh....

Not quite that simple.

As you've touched on there are shades of grey on what constitutes original content.

100% original won't always win against 10% non original. Link juice is a big factor, but overall it's always best to serve up 100% unique original content.