Duplicate content at Search Engine Strategies San Francisco
Shari Thurrow, Founder and SEO Director at Omni Marketing Interactive is up again
Ok let’s get down to business: Search Engines hate duplicate content, because users hate duplicate content.
Website content should be findable via:
- Browsing (from the home pages)
- Querying (search engines, social sites, video sites)
- Asking (refers, friends, family)
When you put too many links on the page, the user is faced with too much choice and the important pages are hard to find. If you don’t put enough pages, search engines filter out duplicate content at multiple places through the crawling and indexing process. They filter out URLs at the shingles (word sets), if the word sets are the same on multiple pages they will be considered duplicate content – even if it’s out of order or totally scrambled.
Dealing with duplicate content:
- IA – make sure of your site navigation
- Robots.txt – are you preventing your site from being crawled? Shari was consulting on a job where the dev team was laid off, the last thing they did was to no index the entire site before taking redundancy – LOL.
- Robots meta tags
- Canonical tags – Shari recons it sucks. She thinks that the reasons the search engines want you to use the canonical tags is because they don’t know what to do with the PageRank – interesting idea. I think both have their place.
Up next is Kathleen Pitcher from Pogo.com about her experience with Duplicate Content
Sometimes, when you duplicate a particular item it can be a good thing (for example if you walk into a cup cake store you can’t just walk out with one – all the pretty options make you want to buy lots of them!)
Pogo.com is an online gaming site. The big question: is duplicate content good or bad for search engine optimisation?
There is naughty duplicate content, or nice duplicate content.
Examples of nice duplicate content include:
- Blog RSS
- Canonical issues
- Print versions
- Multiple products that belong to different categories
These are all legitimate (KG: but still possibly harming) examples of duplicate content.
Naughty duplicate content is black hat, for example taking the same content and spinning it out across multiple domains hoping to get some value from it all. Or even stealing content from another site and putting it up as if it is your own.
If you have duplicate content you might see some of the following symptoms:
- Visibility decrease
- Less desired version ranking
- Content filtered
- Content based algorithmic factors
Examples of Pogo duplicate content:
Pogo had duplicate content, particularly games appearing at different locations. They prevented this by means of user agent detection: if a bot appeared they sent it the clean URL, and if a user appeared they sent the URL with the tracking code. Pogo also used canonical meta tags to ‘fess up’ to search engines about their duplicate content.
Pogo also had loads of user pages where people didn’t enter their unique information; the only difference on thousands of pages was the avatar photo. In the end to remove all these highly similar pages, they put a block in the robots.txt file.
Pogo also had multiple games that were very similar pages. The games pages didn’t rank properly because they didn’t have proper descriptions. The solution: Pogo rolled up their sleeves and wrote unique pages of content for each of the games pages. And the results: excellent. 198% increase in traffic for the term ‘online games’.
Kathleen’s best practices
Do you have duplicate content?
- Ask yourself is each page valuable?
- Is it naughty or nice?
- Talk to your agency
- Talk to your departments
- Review webmaster forums
- Talk to industry peers
- Write unique page content
- If you have duplicate content then find your authority pages
Manage syndicated content effectively
- Give yourself lots of time to get your content indexed before your affiliates
- Make sure you don’t let syndicators use your keyword rich meta data
Michael Gray of Graywolf
Michael starts out by saying that if you can’t get a link to your original content from your syndicator don’t worry too much. My ear’s prick up because this is not what I have heard from other trusted SEOs. You can link back to them and it will be ok, provided they are a more trustworthy site than you are.
If your content is on another site, just change the title and re-write a small section to get credit for both lots of content.
Graywolf’s black hat techniques:
- It is possible to outrank someone for their own content: just grab their content, point links to it and make it more trusted than the original source
- Most web scrapers are stupid: they leave the links in the content. As the little links from scrapers add up they start to count. Support them with good quality links. Mix up the anchor text and surrounding content.
That’s Michael all done, short and sweet.