If you’re using Time on Page, to evaluate content you’re doing it wrong
Ben Young | January 11, 2016
Time on Page, measures when you arrive on a page and when you click off, the difference between those two.
But what happens is that with a bounce rate of 30-40%, or even higher on amplified content (70-80%) that calculation is done off a small percentage of the visitors to the content.
i.e. take this example, we amplified our post The Best Native Ads of 2015 with Yahoo Gemini.
The top url, is our piece of content, the second is the amplified url. You can see from the above data, that the base content got 2 minutes, 20 seconds and the Yahoo Gemini traffic had an average of 5 minutes, 8 seconds.
BUT, if you glance over to Bounce Rate, you see 86.44% of people didn’t click on another link for the baseline content and 91.41% for the Gemini traffic.
It doesn’t make sense, that the amplified source, drove twice as much attention as our organic sources does it?
Google Analytics reports this incorrectly, because it is taking the sources it has data on and extrapolating that.
So, of the 128 visitors from Gemini, Google only has data on 11 of them, as 117 (128*91.41%) left before Google collected any data.
So from 128 visitors, only 11 had recorded data, for an average of 5 minutes, 8 seconds.
Even this number if conflated, due to people leaving tabs open, revisiting content.
So that’s why with Nudge, we measure Attention Minutes, the active time on the content.
From Nudge data, we know that the Yahoo Gemini traffic is on average driving 0.2 Attention Minutes.
How this works is, Nudge measures the active time on the content, whether or not you click any links. Meaning we capture ALL of the attention. Our dataset is each and every visitor, 1 in 9, as GA was collecting.
Further, if you scan over to Bounce Rate, we’ve updated the metric, so that if someone reads the content, they don’t count as a bounce (makes sense huh), there we can see a much better number, 1 in 3 visitors from Gemini didn’t like what they saw so left without reading, a little high but a lot more acceptable than 91%.
In summary, for content marketing, time on page doesn’t make sense, what you care about is if people read your content – and for that Attention Minutes is a great solution.