Syndicating content is an important business development initiative for publishers; it generates revenue, increases exposure, drives traffic and helps facilitate inbound links. However from an SEO perspective there is a downside as syndication creates duplicate content issues.
Search engines don’t want to show users multiple versions of the same content so when an article has been syndicated it is likely that one version will be given prominence – and that may or may not be the original.
One of the most common concerns I hear from publishers is the fact that syndication partners are outranking them for their own content. This happens fairly often, especially when the partners are strong, authoritative domains and their syndicated versions attract a lot of links or social signals.
To some extent publishers have to accept that they can’t have their cake and eat it too. If you are going to license your content to other sites there is always a chance that those sites might outrank you for that content.
The only way to completely eliminate the issue is to require syndication partners to block their versions from the engines. Partners typically refuse such an arrangement, but it is becoming a more common request in contract negotiations.
Beyond that, there are steps that publishers can take to help reduce the risk of being outranked for their own content:
- Require partners to link back to the original on every syndicated article, for example: This article originally appeared on Example.com: [direct link, ideally with the headline as the link text]. It is important for the link to point directly to the original URL
- Publish the content on your site and allow it to be indexed prior to releasing it to partners
- Limit the amount of the text that is syndicated – instead of giving partners the full content, allow them to publish a reduced snippet of the article
- Require partners to use generic title tags (e.g. their site name) on their versions
In its tips for dealing with duplicate content, Google specifically refers to the attribution link but with a caveat:
Syndicate carefully: If you syndicate your content on other sites, make sure they include a link back to the original article on each syndicated article. Even with that, note that we’ll always show the (unblocked) version we think is most appropriate for users in each given search, which may or may not be the version you’d prefer.
You can see that the attribution link is recommended but it is not guaranteed to resolve the issue. So it is important to build in as many protective steps as possible into any syndication contract.
UPDATE:
In December 2009 Google began supporting the rel=”canonical” tag across different domains, giving publishers another tool to use in mitigating duplicate content issues caused by syndication. Check out Will Publishers Add Cross-Domain Rel=Canonical to Syndication Deals? for more information.
Rahul says
Hi Adam,
Great post ! From a publisher perspective if we give a live link (do follow) to all articles/news websites.Then don’t you think we will pass our own link juice (Value) to those websites. Giving away 1/2 links is ok, but when you are syndicating content through out year, you will end up with thousands of out bound links (one way )!! Don’t you think it will de-stabilize the whole website architecture in google and then you may loose everything things!!
** assuming use of no follow is out of question.As google strictly say – you must have to give a do follow link if you are syndicating content from other website. ***
Content Syndication is like a double edge sword. It is going to hurt you from either side. What you say?
Regards
Rahul
Adam Sherk says
Thanks for your comment Rahul. I agree that there are SEO implications that both content creators and syndicators need to consider. But since syndication is an important source of revenue for many publishers, and there will always be sites looking to add good content that they may not be capable of creating themselves, I expect it will continue to be a common practice. So it’s important for both parties to be aware of the SEO implications and to try to mitigate any negatives as best they can.
Alex Kahl says
Hello Adam,
Thanks for the advice about duplicate content. I agree that attribution links are vital, and I recommend not posting all the content
Do you have any information on how to tell when an article is considered duplicate? In other words how different does text have to be to be considered different?
The good folks at google state “Duplicate content generally refers to substantive blocks of content within or across domains that either completely match other content or are appreciably similar.”
What if 5 out of 500 words are different? 5 out of 50? Is being “appreciably similar” defined by some mathematical formula – say for example “the percentage of words that are the same in a given text exceeds 75%?”
On a related note, if an article or press release is supposed to be posted on several locations, what’s are some rules of thumb for rewriting the text in such a way to avoid it from being considered duplicate content?
Adam Sherk says
Good questions Alex.
The engines avoid providing percentages or specific formulas for what constitutes duplicate content. This is in their best interest, as revealing that information would make it harder to prevent abuses.
I think a good rule of thumb is, ‘Would users consider these two pages to be essentially the same article?’ If so, there’s a good chance that the engines will too.
In terms of rewriting or reworking content, a good approach is to summarize key points and/or quote select passages and then add your own take or spin to it. In this way you are adding value for users, which in turn makes the content more valuable and less likely to be filtered as duplicate.
Shailendra Mishra says
Hello Adam,
I know I am bumping a very old post of your blog, but I can’t help that, I read it just now.
In your last reply you mentioned “Would users consider these two pages to be essentially the same article? If so, there’s a good chance that the engines will too.”
Here my curiosity lies in the fact that how search engines can possibly distinguish two articles to be similar or not similar if they are not exactly the same. I mean how much intelligent the search engines can be to decide whether an article is rewrite of a given article or not.
Adam Sherk says
Hi Shailendra – that’s the nice thing about search; older content gets to live on as new people find it over time.
Actually the engines are fairly good at identifying instances of partial duplicate content too. It’s not too difficult to determine if various chunks of text have been reordered on a page, or if X% of the content between two pages is the same.
Here’s some complicated stuff from Stanford you might be interested in:
http://nlp.stanford.edu/IR-book/html/htmledition/near-duplicates-and-shingling-1.html
This topic was also touched on at an SES session this year:
http://www.seroundtable.com/dup-content-sessanfran2011-13832.html
And for some more recent information on duplicate content I’d recommend:
http://www.seomoz.org/blog/duplicate-content-in-a-post-panda-world
Shailendra Mishra says
Thank you for sharing even more material on this topic. Having a quick read of articles you mentioned, I understand that finding ‘nearly’ duplicate content is possible for search engine bots.
Here I want to ask one more question, that, several of so called SEO experts distribute 1000s of copies of an article, rewritten (by hand or by using some software) at word level or sentence level or paragraph level into 1000 different articles and submit them to different websites. Most of the SEO services being sold in the markets involve submitting many copies of rewritten articles with a backlink inside. If Google is smart enough to sort out these practices, why do they are encouraged by improvements in SERP’s and many of their backlink pages being indexed by google.
Adam Sherk says
You’re right, those types of “article spinning” tactics are unfortunately pretty widespread. Google has certainly been making efforts to crack down on things like spinning and scraping, so hopefully we’ll see less of it in the future. In my opinion it is something to stay far away from.
Also keep in mind that just because links exist somewhere doesn’t meant that those links have SEO value. It is always possible that Google will disregard or devalue links that appear to be manipulative.
Shailendra Mishra says
Thank you for your replies, I am looking forward to read your blog regularly.
Good Luck. : )
James Sandberg says
Wondering if syndication is still a good idea post the panda update. good article… can you let me know your thoughts on this?
Adam Sherk says
Good question James. Apologies for using the “Q” word, but it does come down to quality. Syndication of quality content between reputable partners is still a worthwhile initiative for business and marketing reasons, and the SEO negatives can be offset to some degree through adherence to syndication best practices. But you make a good point that care needs to be taken.
Joshua S says
Great article. Adam, lets say the original article contains 2 anchor links. Now the article is syndicated however this time the anchor text links are changed but The syndicated article point contains another link pointing to the original one.
Is this duplication or syndication?
Adam Sherk says
Thanks Joshua. If the only difference between the original and the syndicated version is that the anchor text of a couple links is different, that is still duplicate content.
Alex says
Hi Adam,
Great post! Quick question, you mention “Require partners to link back to the original on every syndicated article, for example: This article originally appeared on Example.com: [direct link, ideally with the headline as the link text]. It is important for the link to point directly to the original URL”.
For SEO purposes, does it matter if this link is at TOP or BOTTOM of the syndicated article? I am negotiating an arrangement with a publisher, and they are opposed to putting a link / sentence like that at the top of the piece…. I want to know if it will be a big deal to have it at bottom. Thanks so much!
– Alex
Adam Sherk says
Hi Alex – the location on the page won’t matter, the presence of the link is what’s most important. The bottom is a more typical location simply because partners usually want it there to avoid distracting their own users.
John D. says
Hi Adam,
Thanks for responding to comments in this post after so many years.
I’m wondering what your take is on “embargoing” syndicated content. For example, is there an ideal time to delay the release of syndicated content (2, 8, 12, 24 hours, etc.) so that Google may credit the original source? Or, once the initial article is indexed, does Google make the same attempt to ascertain the original publisher regardless of when the syndicated article is published?
Thanks in advance,
John
Adam Sherk says
Sure, John, happy to help. Generally speaking delaying release to syndication partners is a helpful tactic, as it allows the original not only to be indexed first but to begin establishing some link and social signals too. There aren’t any hard rules on how long to delay, but roughly speaking the longer the better.
With news content that has a shorter shelf-life you typically can’t delay by a day or more, but even a matter of hours can help the original to get a head start, particularly for Google News. For evergreen, non-news content longer delays are usually possible since the content itself is not time sensitive.
John D. says
Thanks for the quick response, Adam — this reaffirms what I had believed.
And thanks again for keeping these old threads alive, it’s one of many reasons why I’m such a fan of your site.
Happy Friday,
John
Adam Sherk says
No problem! And thanks John, I appreciate that.