The Future of Across-Screen Advertising: Part III

How to Activate the Social Alertness Effect to Raise Ratings and ROI

In this series I am providing more in-depth coverage of the information contained in the CBS presentation that Dave Poltrack and I delivered at the recent ARF Audience Measurement Conference.

In the prior post I reported that separate studies by Turner and Symphony Advanced Media, using vastly different methodologies, have discovered an increase in advertising effectiveness caused by electronic socializing while watching TV. My hypothesis as explained in that post is that while in touch with other human beings we become more present and alert, and this rubs off on attention to the TV screen content whether programming or advertising at the moment.

SymphonyAM found tightly-controlled advertising recall could be increased up to 6 absolute percentage points or up to 100% in terms of relative percent increase. Turner (Jack Wakshlag) found a 30% increase in biometric/eyes-on engagement during electronic socializing concurrent with the same brand’s ad on TV and mobile exposed simultaneously. Turner also found that when the ad is on both screens at once (not necessarily identical content) the increase in brand favorability can be as much as +132%.

Now that results have been replicated, it is time for the marketing innovators to make use of the Alertness Effect to increase the stickiness of their own shows and the ROI of their screen ads/custom content. This post will focus on the practical steps one must take to pragmatically monetize the Alertness Effect to one’s own advantage.

The industry’s systems have been built for Nielsen sex/age CPMs, with campaign sex/age reach to be the only non-judgment factor used in decision making (sometime sex/age/income). Today those systems are beginning to be joined by new programmatic systems with faster and potentially more multidimensional signals used to make decisions, allowing for feedback loops to improve TV buys in flight as is already being done in digital.

These additional potential signals can include real sales, URL visits, tuning duration increases, data on Alertness Effects and any other types of data that is found to be predictive of ROI. We have highly recommended the practice of calibrating proxy metrics used in targeting and media selection, to ROI. TiVo Research/TRA has led the field in actually doing this, finding their purchaser target concentration score the True Target Index (TTI) calibrates 0.7 to ROI, meaning a 10 point increase in TTI on average results in a 7% increase in ROI. The same thing should be done for all data used in targeting and media selection for all media.

Even in the more tightly constrained environment of the legacy systems for turning data into decisions and transactions, there is a single column set aside in Mediaocean where any number of signals can be combined into a single adjustment factor to apply to the Nielsen sex/age impressions to derive an audience size metric that is weighted to reflect, for example, Alertness Effect.

The steps therefore for the media practitioner are as follows to increase ratings and ROI:

  • Obtain a source of data on the degree of Socialtasking (the SymphonyAM term for electronic socializing) going on concurrently with the viewing of each “buyable” TV show. This is a plus sign for a program.
  • Same for Gametasking (SymphonyAM term for playing electronic games) going on concurrently with the viewing of each “buyable” TV show. (This is shown by SymphonyAM to be a detractor from advertising recall.)
  • Feed these data into the buying equation whether in Mediaocean or any other system. Presumably one has already been using purchaser target information as a weighting factor, probably as an index; weight these new attentional factors into this index in order to have a single “quality”/“value” variable so that workflow remains as before. Purchaser targeting is expected to have higher ROI lift than attentional factors, but we could be surprised.
  • Apply them to the Nielsen impressions positively in the case of Socialtasking and negatively in the case of Gametasking.
  • Do everything else the same as before. This will lead to a different choice of programs (or network/dayparts in the case of rotation buying) that will result in higher ROI.
  • For a “tune-in” advertiser the result will be higher ratings for the promoted properties/events.
  • Look for, request and explore opportunities to present simultaneous brand and/or program content across two screens. Test this and learn how to do it better and better over time with efficient ongoing built-in research feedback loops.

Over time, ask your researchers to:

  • Tune the equation by calibrating Socialtasking and Gametasking to ROI/ratings lift the way TiVo Research/TRA calibrated its TTI to ROI. This will further increase the lifts you get in ROI/ratings. My hypothesis is that this metrical combination will have a 0.4 calibration to ROI/ratings lift. Meaning that a 10% increase in the weighted audience “value number” will average a 4% increase in ROI/rating.
  • Keep an eye out for other new metrics that can help better tune the buying equation. There is no reason to believe that simple audience size is the dominant factor in the equation of advertising effectiveness. We might find in the future that the emotional metrics, which we have never developed adequately before, have a combined influence on ROI that is greater than audience size per se. That could lead us to break the pattern of acting as if one data source is the currency when actually it is one of the factors that are needed for a currency but not in itself the entirety.

In the fourth post of this series we will bring to light another important emotional dimension that can be programmatically included in the perfected ROI-based buying equation of the present and future. 


- Bill Harvey is a well-known media researcher and inventor who co-founded TRA, Inc. and is itsBill Harvey Strategic Advisor.