Everyone’s chasing real-time analytics, but what are they learning, and what is it doing to journalism?

Big Data is a lot like a teenager — growing fast, but so very misunderstood. What’s more, the lack of serious understanding is leading to a series of very odd strategies that may suggest why the public has such a low opinion of journalism at present.

Ever since it became the thing to talk about in the early part of this decade, every online company worth its salt has to have a Big Data team, or at least be willing to show that they’re on top of their numbers.

Dejan Nikolic discussing editorial analytics and Content Insights
Content Insights: changing the way we approach editorial analytics

There’s a problem that few of them will admit to, however. While everyone has their analytics packages plugged in, knowing what to do with the data they’re collecting — how to read it or how to act on their findings — is a whole different ball game. Even in cases where the company has hired a data analyst or ‘scientist’, knowing what questions they should be asking seems to be a conundrum they’ve yet to address.

In the publishing industry, this is especially obvious. We set up Content Insights because we were editors unsure of the questions we should be asking, and unsure of what the numbers we were looking at really meant. As we’ve pointed out elsewhere on this blog, tools such as Google Analytics and Omniture were set up to serve marketers rather than journalists, so discovering precisely how engaged your audience is with the content you’re creating is almost impossible. Content Insights was created to reverse that situation, and to present the data in a way that suited people who don’t hold degrees in advanced mathematics.

We’re not the first, of course, but increasingly we’ve noticed that other tools created for the editorial community share one thing in common that we’re not entirely in agreement with. Invariably, our competitors highlight the importance of real-time analytics, which — to a journalist — is like a cross between seeing their name up in Beatles-sized lights whilst at the same time being offered a free smorgasbord of class-A drugs. Sure, they get to delight in how wonderfully their latest piece is doing, but that horribly addictive adrenalin hit will be short-lived and leaves them wondering which of their journalistic values they have to flog in order to get more of the same.

Real-time analytics are like class-A drugs to journalists, leaving them wondering which of their journalistic values they have to flog in order to get more of the same

Sure, there are things that real-time analytics can tell you, but they tend to confirm a strategy rather than inform it. This ‘confirmation bias’, as we call it, shows only that what you’ve created has had some form of instant effect, and highlights the platform that your audience tends to be arriving from — both of interest in the short-term, but unlikely to give you much in the way of insight that could be used to build a longterm strategy.

What real-time analytics do very well is encourage the pursuit of clickbait. Take, for example, The Guardian’s coverage of Manchester Utd’s recent defeat at Watford — a game that most pundits felt they should have won with relative ease. In the few days since the game, the website had published no less than eight articles on the same subject, trumpeting words like “crisis”, “stop the rot” and “Mourinho under scrutiny” in a series of headlines that the tabloids would be proud of.

Screen Shot 2016-09-20 at 16.04.25

This is common practice at many leading newspapers, especially those that rely heavily on advertising for funding. The strategy seems to be: watch which articles ‘trend’ in real-time and then pump out five more articles on the same subject, using words that are guaranteed to push the reader’s emotional buttons. It’s a strategy for now, but it’s certainly not a strategy that can be seen as longterm. Below the line in the comments section, the readership seems to have cottoned on to the game that the editors are playing. Poor Jamie Jackson can’t publish an article without someone accusing him of clickbaiting, as if spending the day pumping out articles that appear to deliberately provoke the readership into a new stratosphere of rage-clicking is his own personal preference.

Screen Shot 2016-09-20 at 16.04.03
Screen Shot 2016-09-20 at 17.01.42

And it’s not just The Guardian, either. In an article published this week on Digiday, the digital editor at the NME, Charlotte Gunn, revealed that editorial meetings at this venerable old music magazine go something like this: “The team gathers together to pitch ideas for the day ahead. Opinion pieces, videos, interviews, galleries: Anything goes, providing it has a killer headline.”

The subtext is obvious: “Anything goes, providing people click on the headline.” The quality of the story, the depth and strength of reporting… none of these seem important anymore. The more clicks the better. We want to see those real-time stats flying through the roof! Don’t ask us how it works, but every click cranks the advertiser’s steely smile one notch tighter.

Pew Research Centre Longform

That’s why we’re more focused on longterm editorial analytics. Much as the longform article tends to be the result of concentrated research usually involving a large selection of articles and data sets, we feel that you can learn more from analytics when they take into account larger sets of data, and when that data is interpreted and presented in a way that suggests a variety of easy-to-understand possibilities. Of course, our approach goes hand in hand with the principles of the Attention Movement, but we’re fine with that. We think it’s a step towards a less frenetic and less indignant world, and it seems to be increasingly what readerships want. As a recent report by the Pew Research Centre suggests, people are increasingly willing to engage with longform articles online these days, even on mobile.

And so we urge a more considered approach that takes into account the relationship between author, audience, referral platforms, topics and sections over a larger period in order to see which areas of content your readership regularly engages with. This encourages strategic thinking rather than clickbait chasing — gaining and keeping your audience’s attention with content that your writers have had time to work on properly, rather than employing a scattergun approach kills audience trust and overworked journalists in equal measure.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.