Is time spent a better metric than pageviews? Upworthy says it is, and open-sources its code for attention minutes


At this point, almost everyone — online publishers and advertisers alike — agrees that raw pageviews are a poor measure of the value that a media outlet provides, but no one can figure out exactly what to replace them with. Some sites have chosen to focus on social sharing, while others are creating their own ways of measuring the actual time that a reader spends with a site’s content. Upworthy’s version of this metric is called “attention minutes,” and the company said Monday it has open-sourced the code so others can use it.

The need for a solid metric is fairly obvious: publishers of all kinds mostly rely on advertising, so they have to be able to show the brands, agencies and networks they deal with that they are reaching large numbers of the people they’re being paid to reach. The easiest ways to do that are pageviews and unique visitors, but both of these metrics are flawed to some extent — and so are raw clicks, likes and shares, as Upworthy points out in its blog post:

“Clicks and pageviews, long the industry standards, are drastically ill-equipped for the job. Even the share isn’t a surefire measure that the user has spent any time engaging with the content itself. It’s in everyone’s interest — from publishers to readers to advertisers — to move to a metric that more fully measures attention spent consuming content. In other words, the best way to answer the question is to measure what happens between the click and the share.”

Many people see social as the new SEO, but focusing on shares is flawed because — as Chartbeat co-founder and CEO Tony Haile has pointed out — the data shows virtually no correlation whatsoever between whether people share a link to a piece of content and whether they have actually read it, since many users are apparently happy to share links to things they haven’t spent any time with at all.

Trying to measure actual reader attention

So Upworthy and Chartbeat both have their own metrics: Upworthy calls its version “attention minutes” and Chartbeat calls its measurement “engaged time.” Although they use somewhat different methods, both track how far readers get through a page of content or a video before they click away, and use other signals to determine whether a page is simply open in a tab or whether the reader is actually involved in reading or watching the content. Interestingly enough, Upworthy says that Twitter does very well as a source when measured by total visitors, but somewhat less so when measured by actual engaged time or attention minutes.

Upworthy attention minutes

Other sites such as YouTube, Medium and the Financial Times also focus on total time spent rather than just measuring page loads or unique visitors — and Medium said recently that it has even started compensating some of its writers based on the amount of time readers spend with their story. But not everyone believes that “time spent” is an effective metric: Gawker editorial director Joel Johnson, for example, told BuzzFeed recently that he would rather measure user satisfaction rather than just the amount of time they spent on a page:

“Perhaps someday, but that’s cart before horse, really; what we want to measure is user engagement through satisfaction. Maybe time-on-page will be part of that, maybe not.”

A Vox Media spokesperson told BuzzFeed that different types of content serve different purposes, and some may not be designed to hold a reader’s attention for a long time. “We don’t really care if someone spends five minutes on a TV schedule, game-time post because that isn’t the point… so time on site as an end-all, be-all metric doesn’t really work at the moment.” Haile, however, pointed out that the service’s “engaged time” is an aggregate over a period of a week or month, so posts that take a short time or a long time both have a place.

Post and thumbnail images courtesy of Shutterstock / ollyy

You're subscribed! If you like, you can update your settings


Comments have been disabled for this post