The Beauty and Risk of Crowdsourced Content: Amazon Partners with Wikipedia

Earlier this week, Amazon began A/B testing a feature that innovatively uses crowdsourced content: instead of using their own, custom author pages, Amazon is using “shopping enabled Wikipedia pages.” They are sacrificing both control and SEO value for the power and robustness of the internet’s most powerful source of social content.

In the past, the search results for an author would include the Amazon Author Page, usually #2 in the results set. Here is a search for science fiction author Glen Cook, showing their own Author Page:

This refers to a dynamically created page, hosted by Amazon, that aggregates the author’s books and a small biography (when available). The books aren’t listed in any particular order, and series aren’t grouped or even all present – this is because Amazon is [generally] just running an advanced search, then taking the user to those search results. More popular authors, such as Stephanie Meyer, have curated pages with more extensive custom content, graphics, and expanded biographies. However, because of problems with scalability, there is considerable inconsistency from author to author; the currency of content must also be a heavy drain on Amazon’s editorial staff (does Stephen King have a new book out already??).

Amazon’s alternative strategy to partner with Wikipedia is brilliant: it is content that is robust, current, interactive, and familiar. It takes the strain off of Amazon to provide (and keep up with) author content, because the author’s fans do the work. Here, you can see the new Wikipedia results for the Glen Cook page:

This is a monetary win-win. As shown below, Amazon is scraping Wikipedia and adding carts next to all the available books; they’ve attached to everything from Common Sense to Knives of Desire. The book titles link to another Amazon-enabled Wikipedia page, and the “See buying info” button leads back to the Amazon product page. Amazon makes it easier to find and buy a book, and Wikipedia gets an affiliate fee (usually 15%). This is the new author page for Glen Cook, complete with integrated shopping carts:

Amazon is clearly using an algorithm to match the cited work to their product that best matches, probably some combination of title+author+year+relevancy of surrounding content. Below is the full Wikipedia version of the “product page,” accessed through title links. Note the prominent buying button next to the title, and the carts everywhere other books available through Amazon.

This isn’t without issue, though. The SEO benefits are likely to be limited, considering that Google doesn’t count duplicate content (though I’m not an SEO specialist, so there may be other benefits I don’t know about). Therefore, they are losing the SEO value of their former author pages, clearly willing to sacrifice it for the value (and user experience) of Wikipedia. They are also at the mercy of Wikipedia users, so any problematic pages will be also reflected via Amazon. And there is always the wildcard of user reaction: reading purists may want more reliable content from Amazon.

However, like Amazon, I believe this is a brilliant and strategic use of information resources. They’ve capitalized on the most robust and familiar provider of crowdsourced content, creating pages that are familiar yet helpful to the user, and easy and profitable for them. As a B&N employee, I hope this feature fails the A/B test; as a reader, I can’t wait for the full rollout.