Commentary

No Attention, No Outcomes

I'm writing this for those not fortunate enough to attend the Advertising Research Foundation's recent Audience X Science conference in New York City. You will want to pay attention, because one session in particular -- the ARF's Tracy Adams' and Paul Donato's review of Phase 3 of its attention measurement validation initiative -- is all about it.

You can read my previous commentaries on Phase 1 and Phase 2, but this year's session, dubbed "Attention As Signal, Or Beyond," demonstrated how many in the industry are now using attention metrics to judge relative media audience delivery, quality and ad campaign effectiveness via the many research vendors available.  However, these vendors use different approaches, definitions and techniques.  

advertisement

advertisement

So, what are best practices and the most meaningful application of these resulting attention data indicators across media that advance media planning and possibly buying well beyond commonly used “viewable impressions, sometimes referred to as “raw impressions”?  

Of strategic note to advertisers is that the Media Rating Council's and the Interactive Advertising Bureau's so-called "viewable impressions" are solely device-based measures and therefor represent no real opportunity-to-see.  Moreover, they are often self-reported by digital platforms, unaudited and with ad creative and placement rarely independently verified.

Brand campaign case studies have proven use of attention metrics consistently delivers more cost-effective outcomes for campaigns for any given media budget.  

Phase 3 of the initiative used a complex neutral design examining media channels, platforms and placement.  It involved eight research vendor attention measurement methods and resulting metrics evaluated against campaign brand lift results across four different campaigns using three different media platforms.  

From a big picture perspective the ARF's Donato highlighted, “At the channel-level, the clearest point of agreement across research vendors was that television -- including CTV, linear TV, and YouTube/online video -- consistently drew more attention than social media. This is likely because most measurement methods employ -- directly or indirectly -- some measure of time.”

In the platform-level analysis -- Facebook vs. TikTok -- there was consistent attention within a measurement methodology, but not necessarily across methods.

In the placement-level analysis – feed, reels and stories -- there also was consistency within a measurement methodology but not necessarily across methods.  Prime-time and late night tend to earn the highest attention scores.  

As would be anticipated, the research found attention alone does not predict brand lift, and media weight plays an important role. Amost certainly along with many other media and campaign execution variables.  

The top line implications for advertisers and their agencies are naturally broad but underline the significant value and the complexities of paying special attention to "attention" and its pragmatic application in media planning, brand campaign by brand campaign.  

Based on the limitations of the research design, Donato and Adams offered evidence-based guidance on attention’s role in media planning, together with how to use attention metrics alongside outcomes measures.  

  • Understand what is actually being measured (definitions and derivations) 

  • Use attention to evaluate media channels

  • Compare to company norms. 

  • Stick with the same measurement company over time.

  • Consider the interaction of attention, media weight and outcomes

As Adams emphasized, “The more important distinction may be between attention and inattention. Advertising can work across a range of attention levels, so long as the content is not ignored.”

To which context of the medium and its editorial environment might be added to ensure brand safety protection whatever the lift especially on social media.  

While the ARF must be highly commended for the scope and depth of its three-phased evaluation, perhaps the best prologue in the use of attention metrics comes from Havas Media Network's Jonathan Waite, which should surely be embraced by all CMOs and brand managers:

“If you are still counting raw impressions today, you aren’t a media planner. You are a passenger in a machine that doesn’t care about your brand’s health.”

2 comments about "No Attention, No Outcomes".
Check to receive email when comments are posted.
  1. Ed Papazian from Media Dynamics Inc, April 15, 2026 at 5:08 p.m.

    Interesting, Tony. 

    I think that the ARF is doing some valuable work re attentiveness. My problem with it is about interpretation. By delving deeply into outcomes as well as the tip of the iceberg--initial viewer attention---we are going well beyond any considerations that a TV time buyer or seller can use as pure outcomes--even if measurable---can not be assigned to specific ad exposures in individual TV shows--aka perfect "attribution". Too many other variables are at play--the impact of previous ad exposures, duplication between TV show and network audiences, changes in  product distribution, sales promotional price-off deals, etc. 

    Therefore I believe that we should consider attentiveness in two related but at the same time separate contexts. The first is attention simply as a way to determine who actually watched a TV show--program content as well as commercials. This is the only way we are going to get a handle on "audience" which is what the seller is supposedly selling to the buyer. We don't get that now with the people meter/big data set usage melding system. 

    The next step, obviously, is trying to evaluate TV ad exposures--attentive exposures----as regards their effect on brand awareness, share of market sales lifts, etc. At this point the way advertisers and ad agencies are organized, makes this the domain of the brand managers, agency account management and, naturally, the "creatives".

    By trying to make everyone--media buyers and sellers as well as brand managers, account execs and creatives function as if they are all fully integrated into a single system, we are unintentionally causing confusion. For example 70% of national TV ad dollars are placed in massive upfront futures deals where there is no data on either the"audience" or, naturally, on the ads, which haven't even been developed as yet. So the buyers might anticipate attentive ad exposures by relying on past data--but not outcomes. 

    I realize that making everyone march to the same drummer might be a good thing--but we are a very long way away from that idealistic situation. Why not deal with each issue as it applies to specific functions instead? That way we might accomplish something.


  2. Tony Jarvis from Olympic Media Consultancy, April 15, 2026 at 5:25 p.m.

    Ed:  Sound further insights and considerations as always.  From a media planning and buying perspective, "Attentive audience (actually watched or heard) exposures" should surely be the  basis for all Cross-Media-Measurement as also urged by Jonathan Waite.  It would establish the minimal media metric threshold to achieve any outcome (verus raw impressions) albeit depending on the impact of "many other varibles", as you rightly point out.   

Next story loading loading..