This year, research conducted in partnership with Digiday revealed that publishers consider time-on-page to be the most useful metric for benchmarking content success, followed by pageviews and social shares.
How have publishers’ opinions on the metrics that matter changed over time? Read on to find out in this piece, originally published in 2013.
Last week, Sam Petulla, who writes for Content.ly’s Content Strategist blog, posted an interview with OpenNews Fellow Brian Abelson, who spent a year embedded within The New York Times organization. In the interview, Abelson discusses how The Grey Lady thinks (or doesn’t) about data.
“At The Times, there’s no sense that it’s worth their while to use data to make decisions on editorial content. If there were, it would be in context of putting data in front of the people who can make decisions rather than the data making decisions for itself. There will always be this line between the editor and the information and the data.”
That’s a line we often hear about at Parse.ly, from product teams that want to experiment more, to editors concerned with the concept of algorithms and machines overtaking their editorial instincts. We don’t view the concept as an “either/or” situation though; we find that editors that use all of the information available to them: data, experiments, and journalistic experience get the most out of all of their whole toolkit.
The trouble with pageviews
What metrics matter for those high-performing editors? Pageviews? Abelson has some thoughts on that as well. A few months back Abelson posted, “Whither the Pageview Apocalypse?” Though some took the post as the final proclamation of the “death of the pageview,” I’m not sure they read the whole piece. Abelson even has to point this out in the comments section: “the point of my post was to reflect on the rhetoric around metrics, not the metrics themselves.”
I asked Petulla for his take away from his talk with Abelson.
“One of Abelson’s most important points in his conversation with us was the level of care you need to take in selecting metrics. Jonah Peretti just made the same point in another interview. I think that publishers will find a lot more value in their content as they continue to use data to understand it more deeply.”
Does that mean that pageviews have no value? An easy reaction is to say no, because publishers can bastardize pageviews through linkbait or other means. The Columbia Journalism School’s Tow Center for Digital Journalism presented a conference on Longform, where Max Linksky proclaimed a sentiment that’s been echoed at other industry events “I believe the pageview is the sub-prime mortgages of the internet.”
Selecting metrics in context
If pageviews are dead or sub-prime, then what metric should we measure for? Or more importantly, what metric will help digital media properties understand how to grow their businesses and provide value to their audiences?
Abelson makes a note that the answer isn’t nearly as clear as some analytics vendors would have you believe.
“Having experimented with many “actionable”, rather than “vanity metrics,” I can tell you that their results are often just as murky and misleading.”
Some metrics have more explanatory power than others. Some metrics model more nuanced behavior than others. The pageview is certainly not a nuanced metric, but it is highly explanatory, at least for sites where the metric is not overtly diluted by implementation challenges. For most content sites, pageviews represent views of their content. Is it as nuanced a metric as visit duration or pageviews per visit? No. Does that mean we should ignore pageviews? Of course not.
Ask not what your metrics can do for you, but what you can do with your metrics
Publishers should realize that vendors pushing their holy grail metric are selling the mirage of a holy grail — and that it will dissipate upon closer inspection. At Parse.ly, we think the world of web data is full of metrics, and if we can bring them all together, we can offer the best possible nuanced analysis to our customers. But we are kidding ourselves that we’ll be able to find one metric that works for every site.
So, what metrics matter? Sorry, it was a trick question. The real question is: Once you have access to data, what are you doing with it, and how does it change your behavior? We’ll be posting case-studies and blog posts about how Parse.ly customers answer that challenge in the coming months. Do you have an example? Let us know!