Why Active Reading Time Matters
It’s not often that you have the opportunity to improve upon Google. In the advertising world, Google is the default for almost everything; search, display, video, ad serving, analytics. The products you know as DoubleClick, YouTube, Google Search and Google Analytics are ubiquitous for a reason, they’re damn good. But sometimes damn good isn’t good enough, as we’ve learned here at Pressboard over the past few months.
When it comes to measuring content, Google Analytics (GA) falls surprisingly short. GA does a great job of measuring page views and sessions, important metrics no doubt and relevant if you’re selling banner ads or running an ecomm store. But pageviews don’t tell the whole story when it comes to analyzing content. For content, editorial or branded, you need to dig deeper. Well beyond the pageview into engagement metrics such as time and attention. While GA provides a measurement called Time on Page, it’s not exactly as it sounds (more on that later). It’s not that Google is intentionally misleading us, it’s simply that GA is more like a set of binoculars, when what you need for content measurement is a microscope.
So that’s what we built. A microscope for content measurement. A proprietary measurement software, that leverages Google Analytics strengths while also compensating for its weaknesses, and it’s baked right into our platform. We now not only count the number of people reading the content, but how actively they are reading, how far they are scrolling and what they do after they’re finished. We call it Active Reading Time and the powerful technology is now included as standard reporting in every campaign facilitated through the Pressboard Story Marketplace.
Blame the Bounce
At one time GA’s Average Time on Page metric was quite accurate. The real culprit here is the ever increasing bounce rate on content-focused sites, such as news sites and lifestyle blogs. “Bounces” are visitors that only view one page on a particular site before “bouncing” to another destination. So if you’re on facebook, you see a link to a Buzzfeed story about adorable cats, you go read that post and then head right back to Facebook afterwards, you’ve “Bounced”.
Bounce rates on media sites have gone from 20% of visitors in the early 2000’s to well over 70% of visitors today. That’s what happens when social media becomes the new homepage. A high bounce rate itself isn’t necessarily a bad thing, many of the most successful online media sites are seeing bounce rates well over 80% on their articles. The bad news is that Google Analytics isn’t able to measure the time on page for any of those visitors that bounce. In other words, Google Analytics doesn’t know what 80% of people are doing beyond the basic pageview. In order to accurately measure how long someone is spending with an article you can’t wait until they’ve already left, you need to know what they’re up to the whole time. That’s why we had to improve on Google. We needed technology that would give us insight into a reader’s attention at all times.
Comparing Google Analytics Time on Page to Pressboard’s Active Reading Time
Using Active Reading Time instead of Time on Page can produce vastly different results at times. Here are 3 scenarios where you may see a difference. Let’s call our reader Shawn.
Scenario A: Shawn is reading about a great burger joint on his favorite site, Lifestyle.com. After reading the article he clicks over to a second article on that same site, this one is about an up and coming local band. Next he heads over to a news site, Newsy.com, to see what’s been going on in the election.
Since Shawn visited two pages on Lifestyle.com, GA can calculate the difference between the time stamps recorded on both pages. GA records it as 10:16:20 – 10:15:15 = 1 min and 05 seconds.
Pressboard is checking every second and calculates the same result for Article A: 1 min 05 seconds. All good so far.
It’s when Shawn heads over to Newsy.com where he loses GA. Since Lifestyle.com’s GA isn’t seeing Newsy.com’s time stamp, the calculation can’t be completed and so GA returns nothing.
Pressboard on the other hand has been checking in for activity from Shawn every second, so as soon as he leaves, we know it. Pressboard records the time spent on article B as 1 min 45 seconds.
Scenario B: The next day Shawn is reading about how to care for your pet on Lifestyle.com, which reminds him that his own dog, Simba, needs to go for a walk. He leaves the article open on his laptop and heads out. After a 12 minute jaunt Shawn and Simba return and he finishes reading his article about pet care before checking out a movie review on the same site.
Google doesn’t know that Shawn left and so when he returns and heads over to the movie review article Google calculates the time stamps as usual and reports Article A Time on Page as being 13 minutes and 40 seconds.
Pressboard meanwhile has been checking in every second and noticed that Shawn hadn’t been active; no mouse movements or scrolling, for about 20 seconds. Since he’s no longer actively reading, Pressboard calculates only the active time and reports 1 minute and 10 seconds on Article A. When Shawn comes back Pressboard will start calculating his active time on the movie review article.
If you relied purely on GA you’d have to assume that Shawn was a pretty slow reader, but you’d have no way to really know what happened.
Scenario C: Shawn is scrolling through his Facebook feed when he comes across an article highlighting amazing local daytrips, sponsored by his favorite car share company. He clicks through to the article and then once he’s finished he heads back to Facebook.
GA cannot record any time on page for this scenario because Shawn is considered a “Bounce”. This is one of the most common scenarios that we see on socially driven traffic. With Facebook representing over 50% of all traffic to sponsored content, that’s a big reason for the huge number of bounces.
Fortunately, Pressboard’s been checking in with Shawn the whole time he’s been on the Lifestyle.com daytrip article, so the calculation is easily recorded as 1 minute and 45 seconds.
While there are other factors that can affect attention, these 3 scenarios represent the majority of use cases.
As you can see, Active Reading Time provides a much more accurate and robust measurement than Time on Page, that’s why we knew we had to build it into our platform. Over the next few months, as we are able to analyze more and more data, we will begin to produce benchmarks to help you measure success and provide deeper insights into your campaigns.
If you are interested in learning more about our measurement methodology please reach out to us at firstname.lastname@example.org