Percentage of duplicate content allowable

[sportstvjobs]

Can you have ANY duplicate content on a page or will the page get penalized by Google?

For example if you used a paragraph of Wikipedia content for a definition/description of a medical term, but wrapped it in unique content is that OK or will that land you in the Google / Panda doghouse?

If some level of duplicate content is allowable, is there a general rule of thumb ratio unique-to-duplicate content?

thanks!

[RyanKent]

You can have some duplicate content in the manner you mentioned above. It is a natural and expected part of the internet that existing sources of information will be utilized.

There is not any magic number which says "30% duplication is ok, but 31% is not". Google's algorithms are private and constantly changing. Use good sense to guide you as to whether your page is unique and offers value to users.

[Dan-Petrovic]

We're currently observing a crumbling empire of websites with auto-generated content. Google is somehow able to understand how substantial your content is and devalue the page and even the whole site if it does not meet their criteria. This is especially damaging for sites who have say 10% of great unique content and 90% of their pages are generated via tagging, browsable search and variable driven paragraphs of text.

Having citations is perfectly normal but I would include reference section just in case.

[seoninja20]

We recently launched a large 3500 page website that auto generates a sentence after we plug in statistical data in our database. All pages are relevant to users and provide more value than other results in serps, but i think a penalty is in place that the farmer update may have detected with a sort of auto-penalty against us.

I sent in a reconsideration request last week, the whole project is on hold until we get a response. I'm expecting a generic answer from them.

We are debating on either writing more unique content for every page or entering in more statistical data to run some cool correlations. The statistical data would be 3x more beneficial to the user I feel, but unique content is what Google seeks and a safer bet just to get us indexed properly.

[RyanKent]

We recently launched a large 3500 page website that auto generates a sentence after we plug in statistical data in our database.

So the only unique content is a single sentence?

Within that sentence many of the words would need to be common as well. Consider a simple site that offered the population for any given location. "The population of [California] is [13 million] people."

In the above example only 3 words are unique. Maybe your pages are a bit more elaborate but it seems to me those pages are simply not indexable. What you can do is index the main page where users can enter the location they wish to learn about, but not each possible result (i.e. California).

Either add significantly more content, or only index the main page.

[seonexgen]

It's quite difficult to provide the exact data as Google algorithm is Google's hidden treasure. Better to keep yourself safe by creating completely unique content, Referring to your example of Wikipedia definition, you can add something like " ACCORDING TO WIKIPEDIA ..... " while copying definition or adding reference links while copying any content from other sources.

Remember that Google is  not only giving importance to unique content but it should be of high quality. That means the article should be innovative like a complete new thing & well researched, so it mustn't be of 200 or less words. So Google will compare the quality of the whole article with the copied content & then it'll decide whether it's a duplicate content article or not.

[Dan-Petrovic]

That sounds like something Google will hate by default. Your problem there is page quantity to quality and uniqueness ratio.

[AlanMosley]

I dont believe you have aproblem if you havea bit of duplicate content, google does not penilize you for duplicate content, it just dosent award you points for it.