On Monday, Upworthy released the source code for the system it uses to measure user-engagement, focused around attention minutes. Upworthy made headlines in early February when it announced the shift away from prioritizing pageviews, instead seeking to identify how much attention users are paying to specific pieces of content. As Upworthy points out, the number of times a piece is opened or shared doesn’t always indicate the degree to which users are viewing it in its entirety. Similarly, just because a user clicks on a piece of content doesn’t mean she’s going to read it all the way through.
Attention minutes considers whether users have paused a video, how they move their mouse, and whether the tab containing the content is open, among other details. By combining old-school tactics of measuring unique views and sharing with newer, more attention-focused methods, Upworthy claims that attention minutes is the most comprehensive method for measuring the success of online content. Ed Urgola, head of marketing for Upworthy, says it’s more than just a strategy for measuring content success. It informs everything they do.
“We actually use attention minutes as a core company goal,” Urgola explains. “Everything from our content strategy, to the design and functionality of our site, to the way we speak to our communities on social platforms and via email is all informed in some way by attention minutes.”
Upworthy isn’t the only site obsessed with attention. Chartbeat, a company which collects data on how users interact with online content, is equally, if not more interested in measuring engagement this way. In March, Tony Hail, CEO of Chartbeat, penned a piece for Time entitled “What You Think You Know About the Web is Wrong.” In the post, Hail argues that the click should not be used as a measure of success. Hail writes that we need to look at what happens “between the clicks,” to understand how users are engaging with content. Chartbeat uses a system called “engaged time,” which “measures the amount of time your audience actively consumes your content.” Similar to attention minutes, engaged time looks beyond the number of times users click a piece of content to how frequently they move their cursors while on one page and how likely users are to return to the site where they accessed the content.
Medium is another site devoting itself to moving beyond the click. It’s no surprise that Medium is concerned with how long users spend on each piece of content; each of Medium’s articles provides the reader with the number of minutes it will take the average user to consume them. They look at “total time reading,” or TTR, and refer to it as “the only metric that matters.”
As Upworthy has pointed out, however, measures of success go beyond just the amount time spent consuming a piece of content. Since your browser records a timestamp when you access a page–but does not record a timestamp when you leave the page (only when you access a new page)–numbers derived from TTR can often be inaccurate. That’s why the folks at Upworthy argue attention minutes has it all.
However, not everyone is anxious to jump on the code. On Thursday, BuzzFeed staffer Myles Tanzer published a piece arguing that measuring the success of online content through attention may be dangerous. Tanzer writes that the trend in monitoring user engagement could lead to an “attention-obsessed media,” and that we shouldn’t be so quick to throw away the pageview as a method of measurement. Tanzer worries that writers will be “quick to game the system,” and resort to cheap gimmicks to keep their readers’ attention. Urgola argues such critics have not looked closely enough at the way attention minutes is built. “We think attention-based metrics are a signal of truth, and that’s hard to ignore and even harder to manipulate with tricks,” he says.
This may be the one issue where BuzzFeed is old-fashioned. According to Christina DiRusso, who handles business, tech, and product press for BuzzFeed, the website is mainly concerned with looking at how frequently its content is shared. DiRusso admits that this method is “unlike most [other] media companies.” However, Dao Nguyen, vice president of growth at data science at BuzzFeed, explains that multiple metrics are monitored. According to Nguyen, BuzzFeed does sometimes look at “what percentage of the page the reader has attained in addition to social metrics.”
Upworthy’s attention minutes may not have everyone convinced, but it does have everyone talking. So far the release of the code has inspired Parse.ly, a company that helps websites measure user engagement and site traffic, to start using attention minutes for all of its clients. Clare Carr, director of marketing for Parse.ly, says that she thinks the metric can help them cater to clients with different goals:
We want to provide our publishers with the option of looking at their stories through any of these lenses: the “traditional” pageview metric, attention minutes, and/or reader loyalty to get the most accurate and relevant understanding of their audience. We believe this can help prevent short term tactics like clickbait, etc., because the additional metrics can be used as a balancing effect; for example, seeing a post with a huge pageview number but low attention minutes and return rates could be an indicator that a headline is misleading.
Attention minutes’ mix of old and new metrics may just win over its critics, if Nguyen’s take on BuzzFeed’s methods of measuring user engagement gains traction.
“The great thing about the internet, is that you edit not because you don’t have enough space in your magazine/newspaper,” Nguyen explains. “You edit because [the piece is] boring.”
Fiona Lowenstein is a CJR intern.