Once upon a time, not so very long ago, the idea of having a data analyst sitting alongside a newspaper editor as they went about their work would’ve been absurd. Not only did the two disciplines not belong together, one of them barely existed. Roll forward a few years, however, and you’ll find that the data analyst, or Big Data department, is commonplace in newsrooms across the world, and editors are having to work to understand how editorial analytics can be put to good use in their quest for a successful editorial strategy.
“We offer the modern editor an easy-to-understand data set that can back up their editorial gut instincts when they need support”
So where does that leave the old editorial ‘gut instinct’ on which newspapers once thrived? The truth, as any editor will be able to tell you, is that gut instinct only ever counted for part of it. “You have to remember that, as a print editor, I was all about copy sales,” said Conde Nast‘s Becky Lucas in our interview last year, pointing out that analytics of some form or another have always been part of the job. Similarly, editors that had the luxury of focus group budgets will have been relying on another form of data to help understand their audience better.
At Content Insights, we specialise in analytics for editors. That’s not to say that trained analysts have no place in the newsroom – far from it, as we’ll see below – but it is to say that there are differences in the ways that the two roles look at data. “Until page views get dethroned, I’d say that Google Analytics is useful to editors as well as marketers,” explains Henrik Stahl, working with analytics at Sweden’s Dagens Nyheter and Dagens industri, “but as soon as the Editorial Metric Paradigm Shift is complete, I’d bump the importance for specific editorial analytics up to 10.”
The ‘paradigm shift’ that Henrik speaks of here is the need to look at analytics in the newsroom from a different perspective, recognising that the kind of metrics available to us in Adobe Analytics and GA encourage you to look towards so-called ‘vanity metrics’ in a way that is unhelpful to journalists. Single metrics of this nature are fine for a marketer who is trying to see how far along the sales funnel their customers have travelled, but they don’t do very much when it comes to observing reader behaviour and genuine attention spent. Henrik says that his publications look more towards ‘in-screen analytics’ for this kind of thing, favouring metrics that indicate engagement and traction above other metrics. “Page views have lost most of their initial value,” he says. “In-screen are kind of a clunky metrics as well, but they’re smooth as silk compared to PVs.”
Over at Zetland, a young publishing house in Denmark currently doing battle with the clickbait cacophony that is the modern web experience, Head of Digital Tav Klitgaard explains that making analytics matter in the modern newsroom is as much about changing the way you view their role as it is deciding which metrics to look at. “We’re interested in conversion rates,” he says, sounding more like a marketer than you might expect, “understanding which stories tend to convert ‘leeches’ to ‘members’. We look first and foremost to our paying readers to define successful stories, even though 90% of our readers are ‘leeches’.”
Leeches aside, it’s refreshing to see that there’s a willingness to understand the analytics and put them to good use, although it does seem as though this is the exception rather than the rule. As Ognjen Zelenbabić of CI Labs pointed out last week, in the editorial workplace “many people know how to open an analytics tool and fumble about in it, but far fewer know what to do with the data they uncover”. This is the simple reason for the existence of the Content Insights tool. We understand that editors are often uncomfortable with analytics, mainly because we have been editors ourselves. The CPI that our users work with allows editors a quick and informative snapshot of how their readership is reacting to their decisions – enough information to clue them in as they head into an editorial meeting without having to request a download from an overworked analyst.
Whether or not analytics have a place in the newsroom is by no means a foregone conclusion. There are editors who see their presence as a distraction, preferring to concentrate on the story in hand rather than the number of shares and likes its predecessors have inspired. At Zetland, analytics tend to be kept out of editorial meetings, Tav Klitgaard says. “We try to go over metrics, trends and patterns only once a week, and sum up the bigger picture once every few months. Daily editorial meetings rarely mention analytics.” He thinks there is an obvious place for analytics in the newsroom, but that there’s a lot to be taught. “In my experience, [editorial staff] often need specific guidance as to what to look for, what to disregard, and to be aware of vanity metrics.”
“Facebook and Google aren’t beating us on pure volume only. They know more about our own readers and users than we do. That’s the main problem”
Henrik Stahl thinks there is a ‘crucial’ role for editorial analytics in the newsroom, both in the understanding of the readership and the production of stories. “Facebook and Google aren’t beating us on pure volume only,” he says, “they are mainly beating us on data. They know more about our own readers and users than we do. That’s the main problem. And it’s a problem that can only be solved with additional analytics resources within the editorial entity. All journalists must have a basic understanding of data, how to measure it, how to find patterns, how to dissect it, how to find stories within it.”
All of which sounds like a positive approach to marrying editorial instinct with hard-nosed number crunching, but Tav Klitgaard is quick to point out potential hiccup. “Knowing so much about customer behaviour also implies a risk of giving the customer what they ask for instead of what really solves their needs,” he says. It’s reminiscent of another conundrum facing journalism today: how much should we allow data-driven algorithms decide what we should or shouldn’t read? By focusing on the minutiae of the available data and allowing it to dictate what we write about next, aren’t we in danger of slipping into an ever-shrinking pool of ‘approved’ content? Isn’t it the editor’s job to break us out of our comfort zones every once in a while?
Again, at Content Insights we’ve taken this into account. Our aim is to provide the modern editor with an instant understanding of which sections work well, whether certain writers resonate with your readership on certain subjects, which stories have a tendency to hold your readers’ attention, and which platforms deliver genuinely engaged traffic. What we are resolutely not here to do is to set journalists off on the stressful hunt for more page views. We like to think we offer the editor an easy-to-understand data set that can back up their gut instincts when they need support.
So, do analytics have a place in the newsroom? We think it depends entirely on the kind of publication you want to be. If you’re hoping for single metrics that tickle your advertisers’ vanity spots, then absolutely: wheel in the realtime monitors – hook yourself up to the clickbait machine. If, on the other hand, you’re interested in understanding the genuine engagement between your work and your readership – seeing what it is that your readers come back for again and again, what they share with their communities and what their communities actually respond to – then a holistic overview can be very useful indeed. And we believe that the more readers employ adblocking software and become tired of being gamed by the cynical clickbait world, the more advertisers will have to look towards an attention-based payment model.
Essentially, we believe that Henrik Stahl’s Editorial Metric Paradigm Shift is already well underway.