By Edward Nawotka, Editor-in-Chief
James Robinson, Director of News Analytics at the New York Times, spoke at Digital Book World earlier this week on the use of analytics, metrics and big data in the media business.
The following is an edited transcript of Robinson’s presentation, sometimes paraphrased for clarity:
We look at audience analytics and insights to improve newsroom decision-making. We started the audience analytics team two years ago.
But there are numerous misconceptions about analytics and data.
First are analytics for measurement. Attention minutes, page views, this works when you have a system and model when you know how it works. But in the newsroom, it is focused on insights and understanding, trying to figure out how the system works — it is exploratory.
These systems are different, but are often confused.
How do you measure engagement? It does nothing for me without understanding “how” and audience engages. It is more like measuring “love.”
It is the difference between what we want our audience to do and what they want to do.
We have a ton of habits but rarely consider what these habits are. We refresh our homepage every two hours because we assume people always want fresh content. Do we know this?
At the New York Times, we know our print audience well, but since we work in the digital realm, we tend to look at everything in the context of the print audience.
It is about asking the right questions. For example, when it came to the story “The Jockey,” first question is we asked the editors was who the audience was for the story… They said, initially, “the audience of the New York Times.” But then I asked them to consider these people in more detail: subscribers, tablet readers, sports fans, horse racing fans…
They said, “this story transcends horse racing” but you need to start with horse racing fans. I look at how many people came to our site and read two or more stories on horse racing, and then read “The Jockey.”
It’s not just about numbers, but about quantitative and qualitative research.
Look for leading indicators. Do the gamers at the New York Times read the article the New York Times publishes about games?
Our job is to illuminate the intent of the work they are doing. You need mutual humility: to understand the content producers are not heathens, but also for the analytics guys to ask good questions that allow access to deeper motivations and intentions.