(Image courtesy photoan on Flickr under Creative Commons)

Recently, content farms have been in the news. From Demand Media’s IPO to Google’s impending crackdown on content farms to AOL’s recent decision to increase content quantity and gross margin, content farming is gaining on Farmville in its popularity.

In sharp contrast to the articles journalists have written about content farms (which I imagine resemble what icemen wrote about refrigerators), I wanted to take a look at the issue from a marketing/technology point of view.

In fact, I think content farm is an unfair term – it’s up there with ‘sweatshop’ and ‘death tax.’ For the sake of fair and reasonable discussion, let’s call it Performance Publishing.

What Makes a Content Farm a Content Farm? Or How Does Performance Publishing Differ from Traditional Publishing?

This is a matter of some contention – one man’s content farm is another’s collection of useful articles. Because content farms do not self-identify as such – Demand Media says it “publishes what the world wants to know” – let’s try to establish some criteria that make a content farm a content farm.

Ideas for Articles are Sourced from Search Engine Data

While traditional publishing relies on what a writer or editor thinks would be interesting, performance publishing analyzes search engine data to assess interest in a given topic.

At one level, this is mass content democracy. (Go over to the Google Keyword Tool and play around for a while. Remember to turn exact match on. ) The search engines represent some blend of a collective id and an all-knowing librarian in today’s age. However, Plato, Aristotle, and JD Salinger would’ve rapidly lost out to Justin Bieber and Twilight .

Demand Media has even patented their system for generating article ideas. They look at search volume, CPC of the term, competitiveness, how long the content will continue to draw searches, and a few other factors. (Hat tip to SEO by the Sea for finding the patent.)

Another effect of this search engine emphasis is that performance publishers must produce content that a) aligns with a massive trend in search volume, to take advantage of QDF, or b) create keyword-targeting evergreen content. (Evergreen content is content that will continue to be interesting well into the future, as opposed to news or commentary on current events.)

Margin on Content Production is Measured and Maximized

In many traditional media operations, there is a sort of magical wall between the editorial and advertising sales sides of the house. P&L was measured over the entire publication (or sections of it), and imprecise circulation numbers were given to advertisers.

All of this is changing – it’s just getting started changing, and it’s not nearly finished yet. Advertising is becoming more and more measurable and performance-based every day. As such, the media inventory advertising is served against is becoming more measurable as well. Now P&L can be measured individually by the article.

Now organizations like AOL measure profitability by the article – and demand an aggregate 50% margin across pieces of content by Q2. (SEOs may wonder if there’s some link acquisition goals from highly unique content that mesh with the search engine traffic goals of the less unique, heavily keyword-targeted pieces of content. )

Armies of Freelancers Create Content

Perhaps the biggest difference between a “content farm” and a regular site on the web is who writes the articles – a writer for a performance publishing site is typically a married woman who gets paid, while the writer for a conventional website is typically an unpaid, unmarried man. Or a well-paid professional with a journalism degree, who has presumably worked really hard for a many years to snag that spot writing for a traditional publication.

(Tellingly, a journalist in Bloomberg describes Demand Media as somewhere the writer’s “pedigree is not important.” The implication is that you need a pedigree to write for a traditional publication. It’s questionable whether this is something bad or internet populism at its finest.)

Moreover, the people writing the articles in performance publishing aren’t typically experts in the topic. As such, the articles end up a re-hash of other resources already on the internet.

I wonder if this is sort of false blow against content farms by the traditional media. The traditional media gets stuff wrong all the time. They’re not experts either. And many of the experts writing today on the web are writing to sell you something, or to build an audience before they sell you something. Unbiased content from an expert is a truly rare commodity.

So these are my three criteria of performance publishing/content farms – Freelance writers, Profitability measured and optimized by the article, and topics chosen by search. Now that we have a definition, we can begin some analysis.

But What Does All This Mean?

Traditional Journalism Will Survive, But It Will Have to Reinvent Itself

In a raft of low quality content, high quality content will become invaluable. While long-form, well-written content will continue to be really expensive to make, a preponderance of re-written garbage will make the well researched, well written stories rise to the top. If anything, more Demand Media-esque content will make the New York Times stick out more.

Search Engines Will Probably Not Be Able to Punish Content Farms

I don’t think search engines will be able to effectively punish content farms. While many smart people will tell you the first rule of SEO Club is Don’t Make Google Look Stupid, I don’t think there’s an algorithmic way you can punish low quality content. While I’m sure Google can tell a performance publishing article from a non-performance article (just like it’s easy to tell a site that’s had a professional SEO work on it from a site that hasn’t), I’m not sure they would punish them. Scraper sites are easily bad, but content farms are a tougher nut to crack.

Blekko has gone so far as to ban a number of domains, but a new domain costs $8. Banning domains will just make the same content expand over more domains, and need better strategies to interlink them. That being said, the folks that make search engines are much smarter than me, so we’ll see what they come up with.

Search Engine Algorithms Will Force Conventional Publishing and Performance Publishing to Cooperate

Instead of viewing content farms like Demand Media and high-brow publications like the New Yorker or the New York Times as complete opposites, realize there’s a continuum of publishing quality and search engine content focus.

(As far as the chart itself, it looks to me like the New York Times has put more effort into search traffic than the New Yorker. I do not know if their content costs more. As for AOL, they are in sort of a swath across the middle of the diagram, reflecting the different approaches they assign to different pieces of content. )

If I were a conventional publisher, I would look at a few different ideas:

• Because of search engines, evergreen content can be the gift that keeps on giving!
• Performance Publishing and traditional publishing is not an either/or decision. For instance, AOL is using traditional publishing to draw links and get big brand dollars, while using heavily search optimized pages to draw common queries and serve contextual ads.
• Content that’s high quality (ie can naturally draw good links) and aligned to search queries may prove to be very profitable.
• Curated pages of existing content may prove to be SEO powerhouses.

These are my first thoughts on content farms. In the next few weeks, I’m going to try to replicate some of Demand Media’s technology and try them out on properties I have access to. I will keep you informed of the results, dear readers.

What do you think the future of content production is? High cost? Low cost? All search? No search? Share your ideas…

Popularity: 3% [?]