Why algorithms are the future of publishing, but the UK is behind schedule

12-01-09 Update: Steven Kydd of Demand Media has replied to my questions about when Demand will reach the UK, and the answer is… “soon“. He writes: “Your timing is impeccable — we were having an international expansion meeting when your e-mail came in this morning.

“The business has been growing so fast in the US that we decided to delay our UK launch slightly so we could deal with the hyper-growth. We don’t have any news to share on accepting non-US contributors yet, but we will shortly. Stay tuned”

So the project is behind schedule here because the company’s still dealing with scaling up over there. The people that admire DM – and are scared of them – will have to wait a little longer yet…

Original: SEO-savvy, algorithm-based publishing houses like Demand Media are changing the way online content is commissioned, consumed and monetised. From one-dimensional, one-way text articles led by the editorial whim of a handful fo editorial execs to thousands of videos and articles every day inspired exactly what readers and advertisers want to see.

But for this revolution, the UK may have to wait: Demand was scheduled to launch in the UK in Q409 but it still hasn’t happened. I applied to sign up as a Demand contributor this week only to be told in an automated message that only US-based writers are being considered at the moment. What has held up the launch, I wonder? SVP Steven Kydd told me in May he’s ultimately targeting France, Germany and other developed EU web markets this year and he wants to have “at least 1,000″ paid freelance contributors by May 2010. I’ve contacted Kydd for more details and will update when I can…

Meanwhile, where does the fear of what Demand is doing come from? Do you care where your online content comes from or do you simply just want the information you’re searching for?

Demand generates as many as 4,000 videos and articles a day on eHow, Youtube and elsewhere. Topics are chosen with help from real-time (that phrase again) algorithms that shows what people are searching for, though the process has a human element too.

It rejects entirely the traditional “we know what’s best” paternalistic media attitude – which may help explain why so many people in the traditional media have their knives out for it. Even some digital-minded people see Demand’s SEO-driven model as a race to the bottom in which “quality” content will lose out, whereas others can’t get past the fear that using SEO in editorial decision making must somehow hinder proper publishing, presumably by watering down the tried and trusted scarcity-based publishing model.

It’s worth noting that Demand doesn’t do journalism as we understand it – it doesn’t touch news, for example – though its system is absolutely something news media could learn from. As Prof Jarvis puts it, the “public knows what it wants to know” a damn sight better than any number of reporters. Essentially, do you and your advertisers want people to read your articles or not?

PO’d by such criticism, Demand founder Richard Rosenblatt has issued a “manifesto” on the company’s aims and ethos and he tells Kara of AllthingsD:

“What is wrong with coming up with a way for thousands of writers–who have been laid off, by the way, from news organizations–knowing exactly how much they make, selecting their own topics and publishing when they want? We’re trying to find a new and innovative way to make content”

And here’s what Kydd told me about how the model works:

Articles are checked by freelance copy editors and moderated by a mixture of Demand staff and its sites’ communities. A former exec at 20th Century Fox, Kydd says the days of ‘intelligent people sat around in a room guessing what the public might want’, then spending millions of dollars on a commissioned project, are over. ‘You need to reduce the risk—if you don’t, you could be terribly wrong…’

If “content farms” are the future, newspapers and trad media had better get themselves a plough pretty soon.

8 thoughts on “Why algorithms are the future of publishing, but the UK is behind schedule

  1. Alex Guest


    This is a really interesting area of digital media.

    There are two issues that I see:

    1. the quality of Demand Media’s output is often contested, much as Wikipedia’s was – and still is to some extent.

    2. how does a news organisation use algorithms to determine what will interest readers, when it’s purpose is to break news?

    For me, how these two matters play out will play an important role in how online media shapes up in the coming 12-24 months.


  2. Patrick Smith Post author

    Good questions, Alex…
    Sure, the quality is contested but, like Wikipedia, there is a human editing element too – Demand employs thousands of copy editors. I worry more about how the old model – guessing what might interest people – is going to make any money when other models make more money.

    Papers should be guided by what editors think is important when it comes to traditional news…. but for everything else, why not find out what your readership and advertisers want and give it to them?

    No one is suggesting that 100% of a website’s stories are commissioned by computer – but if you can use technology to find out what issues, trends and stories are interested in it would be foolish not to use it in my view.

  3. Pingback: uberVU - social comments

  4. Pingback: One Man and His Blog

  5. Pingback: psmith, journalist › Book publisher shows HowTo make money from Google ads and free content

  6. Pingback: Curating the web one story at a time » Demand Media – They Get It.

  7. Pingback: psmith, journalist › Exclusive: Demand Media now accepting UK, Canadian freelance writers

  8. Pingback: psmith, journalist › Demand Media: The $114 million content machine that has nothing to do with news

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>