Commentary
The Challenge Of 'Attention' Measurement
- by Tony Jarvis , Op-Ed Contributor, June 7, 2021
It raised some fundamental questions for advertisers, agencies and the media regarding media and ad-campaign terms and definitions, and attempted to address what metric/s should be used as a media trading currency. Or did it?
The star of the webinar was Lumens Research Managing Director Mike Follett, who reviewed the key points of a new booklet being released about how advertising works based on its proprietary eye-tracking studies.
Lumens measures the value of attention utilizing an eyes-on-plus-associated dwell-time method, and I believe has confirmed the answers to some key industry questions. However, I am sure there will be various interpretations of his remarks.
Despite the Advertising Research Foundation’s media model (“Towards Better Media Decisions”) and its crystal-clear hierarchy of ad-campaign effects and their relative incremental values, the U.S. appears to be going in a different direction by embracing a wide variety of nebulous media “impressions” for planning, buying and selling. The impressions gutter, as I call it, includes the so-called “viewable impression” that do not even have an audience exposure component.
advertisement
advertisement
Follett addressed that folly by unequivocally stating that “viewable impressions” are not the same as content “viewed” by an audience and further that they are not equal, equivalent or even correlated.
He did note that in measuring attention to content on a screen it should meet the Media Rating Council's standards for content rendered on a screen, or a “viewable impression.” Perhaps this would be better termed "CRC," for content rendered counts?
Lumens, working with Ebiquity and Dentsu among other major players, has developed a new approach to campaign investment -- “aCPM” or attention cost-per-thousand. Its methodology can determine how many attentive seconds are derived per thousand impressions.
Attention measures have also been shown to relate well to desired brand effects. I suggest this impressions value conversion to “attention seconds” is like a target audience exposure adjustment for a media vehicle (i.e., looking) combined with an impact adjustment i.e., (seeing over time) for the creative execution.
The data shows a relatively small percentage of ads are viewed “with attention” as a percent of the total time media is consumed, and the Lumens data underscores that advertising is a weak but relentless force that builds effects over time. “Like a stalactite,” as Follett remarked.
Based on aCPMs, a campaign's goal should be to lower the aCPM even if traditional CPMs go higher.
TVision CEO Yan Liu echoed Lumens’ findings.
As most readers know -- or should know -- CPM is an acronym that in my opinion actually stands for “completely positively mad.”
Liu essentially confirmed this, noting that TVision has found no correlation between CPM and attention seconds. That's no surprise, as the ARF media model identifies attention as primarily creative-driven, while CPM -- especially when based on nebulous impressions -- is media-driven.
So will attention and its associated aCPM become the new media currency? I suggest that it cannot be a “media” currency, because it is primarily driven by the impact of the creative message, over which media has no control.
Tony, as you may know, I championed the concept of attentiveness many years ago and have kept at it ever since. However, you can't simply assume that all seconds of "attentive" viewing of a commercial are equal and all you have to do is calculate your CPMs based on total attentive seconds, in aggregate, if you wish to get more people exposed to your ad campaign. For one thing, this assumes that the ad sellers are totally passive numbskulls and they won't recognize which programs are in demand due to attentiveness metrics and which aren't. Once that happens the sellers will simply jack up their CPMs for the good shows---probably by a higher amount than is gained in terms of eyeball contacts.
Returning to my main point. TV commercials of various lengths---"15s" and "30s", mainly----tell a story that starts by catching the viewer's attention, quickly gets into its message and finishes---often with a call to action. The question arises, at what point does the viewer---the attentive viewer---get enough of the message and sales pitch for it to be effective?Is one second, enough---a random second? How about two seconds or three? How many viewers who "saw" three seconds of a 30-second commercial got the message? How does this compare to those who "saw" 15 seconds---or 30 seconds? Are all of these seconds of equal value? I say no.
In our haste to get attention metrics into use---which I agree with---we can't forget what the real purpose is---namely, to facilitate ad message communication---without which we are just kidding ourselves--again.
Exactly. However, will Mike Follett and Yan Liu agree with your/our perspective?
The ARF Media Model offers a helix to understand the stage by stage effects of an ad campaign and Communications is identified as Level 5. Certainly not a media currency. Nor, I believe, is "Attention", Level 4, ("seeing") due to the heavy inlfuence of the creative. Which bring us back to Level 3, Ad Exposure to the target group or Eyes-On ("looking"), surely the most valid and valuable media metric to the sellers and the advertisers as stressed by the ARF Model.
A while ago, Josh Chasin and I essentially asked, "When is an impression not an impression? There are many considerations but one answer is crystal clear, "When it's a "viewable impression"!
Until that misguided term is replaced by "Content Rendered Counts", CRC, the fundamental confusion on the relevance and value of various media metrics will continue to proliferate. Consequently advertisers will pay considerable sums for ads often perfectly rendered by a media vehicle that will never be exposed to their target audience .
Bravo Tony and Ed.
First, in Australia CPM is mainly used as a monetary value comparison metric - how many thousands in your target audience 'saw' the ad and how much does that cost (it basically implies that any viewers outside the target have a value of zero - which of course is not universally true).
Most of the neuroscience data I have seen show that effective ads start with a 'hook' ("hey, that product is for me!"), keep you engaged (albeit at a typically slightly lower level), and then finish off with a tag or an offer ("I'll look it up online or in the shops next time").
The key thing is that the media owner has a degree of control up until the ad starts and needs to be held to account (cost, audience and efficiency). Once the ad starts it is all up to the creative then which is out of the media owner's control.