A slow Labor Day news day maybe I thought when I noticed that the NYT was carrying a piece on Monday entitled : ‘Some Newspapers, Tracking Readers Online, Shift Coverage’. As old news goes this is positively antediluvian isn’t it? Despite the fact that most newsrooms have developed metric tracking systems for their websites, and many use them creatively and effectively, it remains a controversial area. Quoted in the piece, Bill Keller, editor of the New York Times says :
“We don’t let metrics dictate our assignments and play….because we believe readers come to us for our judgment, not the judgment of the crowd. We’re not ‘American Idol.’ ”
The following day in the Washington Post Howard Kurtz wrote an article on a similar theme entitled Appeasing the Google Gods. In it Kurtz again outlined the innate newsroom fears of looking at who is reading what as part of the editorial process:
Naturally, those who grew up as analog reporters wonder: Is journalism becoming a popularity contest? Does this mean pieces about celebrity sex tapes will take precedence over corruption in Afghanistan? Why pay for expensive foreign bureaus if they’re not generating enough clicks? Doesn’t all this amount to pandering?
Kurtz, sensibly , concludes that strong news brands are just that, and do not benefit from over-indulging in the chase to the surgically enhanced celebrity bottom. But if arguments are still being had in newsrooms that metrics might actually threaten the quality of journalism, then this is depressing news. The existence and use of metrics strengthens the needs for editorial judgment and moreover an ability to receive, interpret and act on live feedback should be a key requirement for any web editor. We do ourselves no favours by suggesting that we are unable to resist the pull of WAG rhinoplasty when we should be writing about the trade deficit just because Google Trends tells us to.
One of the most interesting projects afoot at the Guardian when I left was the wonderful Zeitgeist, whereby Dan Catt and Meg Pickard initiated work into looking much more closely at what constituted a ‘hot’ Guardian story – explained in more detail here. Part of the curiosity was to look beyond the attention data of ‘most read’ and into a more subtle algorithm of what constituted stories which created a buzz or interest among core readership. The idea that you would not then take the next step and look at how that might inform your commissioning process, must be wrong.
There is nothing to to be afraid of here. And it is not the case that it inevitably produces ‘X Factor’ journalism. Take this week for instance when the slow burn and massively under reported phone hacking scandal caught light through an extensive piece of reporting in the New York Times and further digging from the Guardian. But there was tumbleweed blowing around the story in terms of other press at the beginning of the week. The consistent pressure pushed the story back into Parliament, and nothing about it came from looking at web metric files. However, when I glanced at the ‘most read’ column of stories on the guardian.co.uk front page at the beginning of the week all five were related to the phone hacking. I tabbed to Zeitgeist – which gives a subtler picture of story performance – to see the same pattern. Every top story related to the scandal. I went to the BBC site where their patchy coverage still had Coulson and co occupying the ‘most read’ top slot. Despite the vast majority of the UK press studiedly ignoring the story readers were interested.
If an editor now seizes on a story which is highly complicated, but deeply important, and it is not well read, or read at all, then the information should be used to make sure that story presentation, linking and marketing improves so that it is. Too often the filtering of metrics are left in the hands of the commercial departments, when journalists should be able to use a whole series of tools and feedback data to put important journalism in front of readers.
The salvation of the media as a business lies in part with understanding and reaching audiences more efficiently. I read with interest a blog post earlier this summer from Ethan Zuckerman, entitled ‘what if search drove newspapers?’ about Zoe Fraade-Blanar’s Current project, which is tantamount to a kind of plug-in Demand Media for newsroom desktops. He wrote:
Zoe had invited me to the critique that precedes the final show of these projects. I happened to be in NYC that day, so I went. I had mixed emotions about her piece – on the one hand, she appeared to have reduced the journalistic filtering process to its basest form, connecting supply and demand in a way that reminded me of Demand Media. In my comments on her piece, I wondered whether she was making a political statement, or whether this was a serious tool. She explained to me – and later to Brooke Gladstone at On the Media – that the tool was designed to let serious media outlets cover the stuff they needed to pay the bills so they could cross-subsidize serious investigative reporting.
I wondered if bright students at NYU were thinking about these issues in creative ways, how many newsrooms are giving over resources to similar creative thinking . Asking the Twitter crowd about creative use of metrics brought forth a torrent of good ideas and excellent links. I’m particularly indebted to Ewan McIntosh whose delicious links on metrics are a great resource on the subject.
Understanding what to amplify, what to ignore and how to commercially use what is available in the field of human interest in news is not going to undermine the core principles of journalism or threaten the quality of reporting. Ignorance and fear of numbers, an inability to determine between what’s valuable and what’s not, the blunt application of crude statistics will however kill your brand and distress your newsroom. As Bill Keller says in the NYT piece o the subject, making resource allocation decisions can now be done with evidence-based metrics, but the importance of editorial judgment as a layer on top of that becomes more rather than less acute.
It is a slight digression, but on a similar theme, we have exciting news from the Tow Center this week, with our first report on online metrics and the necessity to identify new standards. A great piece of work by Lucas Kelly and John Graves entitled Confusion Online: Faulty Metrics and the Future of Digital Journalism . Please read and feed back, here is the CJR article relating to the report. I can unfortunately take no credit whatsoever for this, but at the Tow we want to initiate research projects which help the industry and explore solutions to help digital journalism reach its full potential. The report is a great start.
Much more on metrics to come.