10 Charts That Are Changing the Way We Measure ContentBy Sam Petulla April 4th, 2014
In March, the “how to measure content” debate came in like a lion. It came out like a lion on steroids.
Were you following?
1. Chartbeat founder Tony Haile published an epic analysis of Internet reading trends in TIME that garnered about 20,000 shares.
2. Contently’s own Paul Fredrich published a manifesto here on The Content Strategist, calling for death to pageviews and a long life for engaged time metrics.
3. And News Corp’s Raju Narisetti fought back at Poynter, saying pageviews are more than a so-called “vanity metric” and have real value for publishers.
Those were just three of the highlights.
In between, publishers around the web posted some of their best insights into what is and isn’t working when it comes to measuring content. We looked it all over and realized that creating a breakdown would benefit our readers, who might not be following this movement quite as obsessively as we have been.
Without further ado, here are the 10 most important data points from the content measurement debate over the last few months.
1. No one is looking at your native ads.
This one’s a doozy. But as Rahm Emanuel famously said following the financial industry collapse, “A crisis is a terrible thing to waste.” As a whole, content marketers are still beginners at telling stories. It’s their version of the “taste gap” Ira Glass talks about when you are first starting out in media publishing. The 47-point disparity above should be seen as a wake-up call. Readers are still finding a lot of branded content wanting, but that doesn’t mean that it can’t get better or that there aren’t success stories out there.
2. What we engage with can’t be measured just with shares or clicks.
Tony Haile’s number one point in his TIME piece (and the foundation of his business, Chartbeat) is this: To understand your readers, you need to measure how much time they spend with your content. Chartbeat has access to a huge dataset from its publishers network. They came away with the stat above, which found, quite simply, that “most people who click don’t read.”
And as you’ll see in the chart below, Haile came to the same conclusion about shares; in the aggregate of the web, there’s not much correlation between sharing a story and actually reading it. In other words, sharing is triggered by an entirely different neurological chemistry. More on that later.
3. Counter: What we read can sometimes be measured with shares.
As a counterpoint, BuzzFeed’s Jonah Peretti took to Twitter to note that BuzzFeed readers are indeed reading the stories they share, as you’ll see in the chart below.
4. More data ≠ better measurement. But you can blend metrics.
Over at Medium, Pete Davies laid out how Medium measures content, emphasizing the importance of precision in today’s “big data” metrics deluge. Medium starts with a ton of data, then slices it down to score it:
“We measure every user interaction with every post. Most of this is done by periodically recording scroll positions. We pipe this data into our data warehouse, where offline processing aggregates the time spent reading (or our best guess of it): We infer when a reader started reading, when they paused, and when they stopped altogether. The methodology allows us to correct for periods of inactivity (such as having a post open in a different tab, walking the dog, or checking your phone).”
5. Advanced content metrics even matter to Upworthy, the royal highness of A/B testing and headline writing.
Upworthy, which made one of the fastest rises to prominence in the history of media last year, said it is introducing a new metric for quantifying its content, too: Attention Minutes. Attention Minutes is a blended metric that incorporates views, shares, and the total time readers spend viewing or reading a piece. Upworthy explained its decision in a blog post:
“We love thinking this way because it rewards us for sharing content that people really enjoy and find valuable—not just stuff they click on a lot. It may mean that we don’t do quite as well on uniques or pageviews, but that’s a tradeoff we’re happy to make because this is a metric focused on real user satisfaction.”
6. When measuring content, you have to remember that attention matters.
Brian Abelson, a data scientist at Enigma.io and the impresario behind CSV Soundsystem, spent 2013 at The New York Times as an OpenNews fellow, studying the venerated publication’s web data. As the above chart shows, promotion really matters for New York Times stories.
The idea that articles are purely organic, free-floating atomic entities that surface in people’s feeds without any input from promotional tools just did not hold up under his analysis. When articles were put on NYTimes.com or tweeted out from a social media handle, those articles were more likely to reach a lot of people. It seems intuitive, but it’s an important reminder and one of the most comprehensive analyses to date.
7. Clickbait keywords don’t result in reads.
In an earlier post, Chartbeat dove into what people are clicking vs. what people are reading. Unsurprisingly, people are clicking things that sound enticing: what’s big, what’s best, what’s the top, what makes you rich. A lot of those stories don’t live up to the hype, though, and readers move on quickly (likely with a bad taste in their mouths). Instead, readers prefer content that’s substantial and emotionally resonant, as evident by the meaty keywords that trigger the most reads.
8. Love ’em or hate ’em, BuzzFeed’s new quizzes are the runaway winner of 2014.
Some think it’s a sign of the apocalypse, but thus far, BuzzFeed’s quizzes are the runway winner for top overall content format of 2014—at least in terms of engagement. Most of their quizzes garner over a million pageviews. Some have over 40 million shares! That’s no different for their sponsor partners. The above metrics from “How Would You Die in Game of Thrones?” shows the quiz getting over 266,000 total engagement points on Facebook alone. The quiz also has over million views.
9. Certain emotions trigger the most intense sharing responses.
The above graphic, from Fractl’s study on what types of content people share, shows which specific emotional responses correlate with high levels of sharing. The bright sunbursts show the emotions that have the biggest impact.
10. Triggering both the emotional and rational parts of the brain is the holy grail of content success.
The same Fast Company article presented a study from IPA Databank about how emotion impacts campaign effectiveness. The article ultimately found that when rational content is enhanced with emotional appeal, that content performs about 70 percent better. As we wrote about on The Strategist, appealing to multiple thinking styles tends to boost content performance.
Is there anything you think we should add? Let us know @Contently.