“Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.
Today’s column is written by Eric Berry, CEO at TripleLift.
The chasm between effective media buying strategies for TV and digital are often so large that one might be excused for forgetting both are simply attempts to advertise effectively for a brand.
At its core, effective advertising means reaching the right audience with the right message at an appropriate frequency. This aligns with the gross rating point (GRP) methodology for TV, where advertisers target specific audiences with an intended frequency.
When buying digital, however, media buyers often use strict attribution with harsh return-on-investment (ROI) cutoffs. There is an understandable cautiousness when buying digital media because of banner blindness, fraud and other legitimate concerns that largely do not exist for TV, all of which has led to the divergence in strategies.
Notwithstanding certain flaws in TV media buying, the merits of GRP-based buying are such that it drives the media buyer to consider the most cost-effective way to target the highest proportion of its potential customers with the most appropriate frequency and message. Brands look to gauge the effectiveness of these media buys through statistical analysis about improved sales or engagement. The reality of conversion attribution is that individualized conversion attribution is never accurate, and thus a statistical approach for TV reflects the realities of a multiple touchpoint world.
Digital – in stark contrast to television – generally is required to leverage conversion attribution on an individualized basis. Meaningful attribution that actually identifies which ads, among the hundreds of possible touchpoints, contributed to the overall sales funnel for each user, is effectively impossible. Nonetheless, impression- or click-level attribution is still used online.
Viewability today is largely meant to ensure that brands and agencies actually get what they purchased: viewable ads. But it also has the effect of beginning to reframe the question of digital advertising to how many times an ad is being exposed to a target audience. This transformative impact is one of the many reasons viewability vendors have raised significant venture rounds.
The promise of combining viewability with audience will allow for the development of a product that functions similarly to GRP. Media buyers could focus on reach, frequency and message instead of worrying about the nuances that make digital so complex, all normalized by single measurement.
While this may cause headaches upfront, it would also significantly reduce any friction associated with moving TV dollars online. Further improvements to the system could include measures not just of viewability, but of the likelihood of being viewed. Through eye tracking or other studies on a statistical basis, viewability vendors could obviate concerns around banner blindness as well.
Pre-roll units, for example, could be measured directly against outstream units such that each is given the appropriate relative credit rather than simply measuring plays. This could even combine with a per-placement-level score that looks at the size and branding quality of a placement to determine just how effective an ad could be, with the ultimate goal of understanding how impactful different positions and types on placements tend to be and adjusting the score accordingly.
Since digital is much more easily measured than other forms of advertising, many advertisers will continue to require CPA-driven ad campaigns to ensure at least baseline return on ad spend. By creating an effective analog to TV, however, buyers could use an independent verification system and focus on the right metrics. Media across all platforms could be purchased far more effectively with a focus on reach and frequency across a target audience, across all devices.
The promise of a much more palatable media-buying ecosystem tomorrow may finally enable television money to find a home in digital.