They may be awash in metrics, but are most newsrooms really getting the right read on who their readers are?
Some experts question whether media companies are properly analyzing the troves of Web analytics readily at their fingertips.
Programs such as Google Analytics, Adobe Analytics (Omniture) and Chartbeat offer publishers insights on page views, unique visitors, time spent on site, referrals and just about every statistic imaginable. Analytics experts, though, warn that "fly-by-night metrics" are not necessarily conducive to accurately evaluating reader engagement.
That's not to say analytics aren't useful; in fact, experts say, the exact opposite is the case.
Dana Chinn, lecturer of Web analytics at the University of Southern California's Annenberg School for Communication and Journalism, says newsrooms should first determine their long-term content strategy, or value proposition, and find which metrics best measure their audience's engagement.
Say a local media company runs a Miley Cyrus story that generates a million page views via search, but the readers are from outside their market and don't return — that doesn't bolster their brand in their community, Chinn says.
According to Chinn, publishers should favor visits, a metric that Google Analytics considers to contain all the user's actions on a page. If a reader comments on a particular story, subscribes to an email newsletter or shares on social media, Chinn says, that gives editors a more complete picture of their readership and what subtopics generate interaction.
"We're not saying that news should be driven by metrics," says Chinn, also a strategist for USC's Media Impact Project, a project co-funded by the Bill & Melinda Gates and John S. and James L. Knight foundations to measure best practices in media analytics. "We're saying use the metrics as a tool to make a decision about what's going to build your audience long-term."
Chinn points to The New York Times as a company to watch. In April, the Times established an analytics team to help its editorial staff make "data-driven decisions" based on how readers interact with their content.
"When you're spending so much time [and] you're putting so many resources into these projects and into these pieces, you really shouldn't be publishing blind," Aron Pilhofer, architect of the Times' analytics team, told Journalism.co.uk in April.
Elsewhere, Chinn says, publishers — who previously only ever had to consider one statistic: circulation — have yet to think beyond "misused metrics" such as page views and unique visitors. Unique visitors over-represent consumers with multiple devices and browsers, she warns, and underrepresent groups of readers who access a site from a shared computer.
"Just because you can measure something doesn't mean you should use the data for decision-making," Chinn says. "There are so many good and strong metrics, so why waste time and resources — and journalists' bandwidth for numbers — on weak metrics?"
Chinn says it's OK if the audience starts small as long as there's growth. Ultimately, she says, the key is to not be a commodity, especially if the publisher plans to charge for its content.
"Good journalism will prevail," Chinn says. "I don't have any doubt in my mind that it will. I don't have the metrics yet to prove it, but I'm pretty confident we're going to get there because media's just like any other business: a quality product for an audience that wants it will be successful. It's just a matter of finding the right way to get there."
Dana Chinn; Todd Cunningham, director of USC's Media Impact Project; James Robinson, director of news analytics at The New York Times; and Daniel Sieberg, a senior marketing manager at Google, will host an hourlong session on analytics in the newsroom at ONA.
