Second Life: How The Story Of Evergreen Publishing Is Being Retold.

Logo of GOOGLE
Facebooktwitterredditlinkedinmail

Republishing, repurposing, reposting, whatever we choose to call it, it’s been taking place for years as savvy publishers make their content dollars stretch further. Relevant articles or pertinent comments often resurface in response to a news item, but there are more general interest features that are non-time specific and these can be used over and over to the delight of publishers.   

Now Swiss news journal Le Temps has begun a Google Digital News-funded project that aims to delight publishers even more. Project Zombie monitors the success of online articles and notifies the editor, via email or Slack bot, which are the most popular and are worth republishing online.

Le Temps publishes 70 articles a day, and so has a database of 100,000 articles from the past 19 years that are currently undergoing classification. Data scientists from the University of Lausanne are using the program to look at what’s being discussed, how well the feature originally performed and what is currently trending in the world. The program analyzes data from Google Analytics and Chartbeat to measure page views, time spent on the page, comments and shares in order to work out a ‘relevancy score’ out of 100 for each news feature. The higher the score, the likelier it is to do well when published again.

It effectively turns the age-old publisher’s ‘finger in the wind’ method into something of a scientific certainty. And it’s not the first project of its kind. Digiday points out that Vox, NPR’s Carebot Project and the Financial Times have all undertaken similar strategies for working out the effectiveness of evergreen content. Project Zombie is the first, though, that promises to pick out high-performing articles in response to what’s trending in real time.

It’s a clever strategy that makes best use of the sophisticated analytical tools and data that are currently available. Publishers taking up the Zombie program will no longer rely on their own memory for features that cover a topic that’s in the news, nor will they have to worry about timing the re-release of general interest features to current events – a helpful ‘zombie’ will send them a good morning email with the information. And as the ‘zombie’ has also assessed which features did well the first time round, the publisher can republish, safe in the knowledge that it is likely to do well again; as data scientist for the project Yannick Rochat puts it: “Giving a second life to those that deserve it.” The only thing that isn’t clear however, is why it’s called Project Zombie.

If you or your zombies want to prolong the lifecycle of your publishing projects, get in touch with Amnet today.