Getting Specific About Tracking Engagement


Getting Specific About Tracking Engagement

By Gibby Miller - Chief Product Officer · October 5th 2016

Over the last several years, we’ve witnessed excitement and debate around the engagement metric “in-view time”, and a shift toward consideration of “time spent” as an important measure of brand exposure for our advertisers. Recent data from an IPG study confirms that the in-view time of an ad is strongly related to ad effectiveness and brand recall. Notably, increasing the in-view time of a display ad by just 6 seconds boosts recall rate by 11%. However, reporting of these metrics can be cloudy, with evolving standards and technical issues related to tracking.

This challenged our team to create a product ... that could provide in-depth and trustworthy reporting.

For marketers, this is a big challenge.  While on one hand we know that time spent with the brand is important and effective, on the other hand the industry struggles with reporting it in a predictable and consistent way. This challenged our team to create a product that not only increased time spent, but to create a platform and methodology that could provide in-depth and trustworthy reporting. We call this platform “Beacon”, and it’s become a central part of our solution.

With Beacon, we can validate in-view time for all of our ad impressions across the entire delivery ecosystem.  Because we often have more information about what’s in-view than our programmatic partners, this allows us to fine-tune and optimize our buys for clients in a better way.  

We begin tracking in-view time of our ad units the moment they meet the MRC standard requirement (50% pixels in view for 1+ seconds), stopping after 60 seconds to ensure we exclude “away from keyboard” data. Once our ad is engaged, we record time spent within the branded content experience for up to 15 minutes, along with scrolling actions, article completion rates, and interactions.

"This allows partners to easily index engagement to spend."

We calculate total time spent by adding the in-view time of the served ad to the amount of time the user spent actively exploring the branded content experience.  Further, we’ve come up with a metric to help our advertisers quantify their audiences’ engagement as it applies to a CPM buy, and we call it “HPM”: Hours (of time spent) per 1000 impressions. This allows partners to easily index engagement to spend.

In-View Time of Ad + Content Experience Time = Time Spent / (Total Impressions/1000) = HPM

In live campaigns, we find our ad products outperform standard display for time spent with brand. In a recent client campaign, we saw an average HPM of roughly 8.21 hours, or 29 seconds per impression on desktop, 28 on mobile.  That’s 4x what the IPG noted as the threshold (7 seconds) for drastic increase in ad and brand recall, and a significant increase over Moat’s in-view benchmarks: 19.2 seconds for desktop, 15.6 for mobile.

The results are exciting. We set out to not only improve engagement and bring clarity and consistency to our reporting, but to lift confidence and excitement about “time spent” as a valuable metric for our advertisers.


Garrett Griebel