"On TV and Video" is a column exploring opportunities and challenges in advanced TV and video.
Today’s column is written by Maggie Zhang, senior vice president of video research and insights at Dentsu Aegis Network.
As the timeless tale of “Goldilocks and the Three Bears” goes, the young girl tastes three different bowls of porridge and finds the one with just right temperature, neither too hot nor too cold.
Today, marketers face a similar task with TV attribution: They must evaluate a lengthy list of attribution capabilities to determine the solution that is “just right.”
The marketplace demands a single, baseline standard across attribution players because the success and continued growth of outcome-based TV performance measurement and transaction depends on it.
Can this demand be met, or is it just a fairy tale?
It’s no secret that the TV ecosystem is undergoing unprecedented change and disruption. With consumers’ behavioral shifts and the growth of addressable technology and data-rich digital video platforms, television, the well-known mass reach and upper funnel branding channel, is under increasing scrutiny.
Marketers want definitive performance proof points and accountability, similar to practices across the digital media landscape. Holistic and unified media activity may sound like buzzwords to some, but in reality, they are not. Marketers want to understand how different media channels impact tangible business outcomes, and TV is no longer an exception.
When we look at TV through a performance lens, TV attribution sounds like an easy sell. For example, a $10 million TV budget used to get us 250 GRPs or 60% reach, and now we can also prove the investment results in 15% increase of incremental website visits or 10,000 units of widgets off the shelf. Not only that, but we can analyze, benchmark, optimize and eventually forecast TV campaign performance by key campaign elements such as dayparts, creatives or tactics.
Armed with attribution capabilities, TV is measurable, accountable and attributable, just like digital media. We can all ride off into the sunset now, right?
Looking for ‘just right’
Not so fast. Imagine a tale with three endings. We tasked three different TV attribution vendors with measuring the same TV ad campaign and each reached vastly different conclusions.
Why were the results so different? What is the source of truth? Who should be the arbiter that determines the true quality of work?
We wanted answers, so we conducted a comprehensive TV attribution vendor audit. We learned that despite the appeal and excitement surrounding TV attribution, it is still in very early stages.
As the quantity of players grows rapidly, so do the variances in capabilities and offerings. There is a disparate array of data sources and inputs, different technology infrastructures and varied levels of methodological sophistication.
Which TV attribution solution is “just right?”
We also learned that marketers’ North Star should always be the brand’s business objectives and campaign KPIs. There must be clear alignment on what the ultimate KPI is and how to measure against it. Deep brand understanding of historical norms and vertical and brand-specific benchmarks are critical to properly interpreting results, even more so when results are counterintuitive or unexpected.
Equally important are quality and representative data sets for both media exposure and conversion metrics, as they are prerequisites for accurate and reliable attribution results. Otherwise, it’s simply garbage in and garbage out.
The third leg of the stool is the modeling approach. There is currently a wide range of attribution methods that run the gamut from control versus exposure lift measurement to statistical multiple regression. Marketers need to feel confident about the modeling approach at the outset and apply it consistently across campaigns over time to gain reliable and actionable insights.
Standardize early for growth
If I could share one cautionary insight from the audit findings, it’s that TV attribution is complex, nuanced and constantly evolving. If not executed properly, it can mislead marketers’ overall media strategy and investment approach, thus dampening adoption at the early stage.
As an industry, the onus is on all of us – regardless of one’s position within the ecosystem – to establish a consistent baseline of standards for this nascent space. While there is a need for brands' nuanced measurement requirements, it is imperative for the baseline attribution methodology and data inputs to be standardized so true progress can be made. For this, collaboration is crucial. There needs to be regular dialogues, debates and audits to ensure quality and integrity of all attribution work. That means brand nuances and caveats can be applied on top of this baseline of standards once determined.
After that, marketers will be able to evaluate attribution solutions using the “just right” principle with peace of mind, just as Goldilocks did with her porridge.
Finally, let’s be clear: This is a challenge we must accept head on. We need to shed legacy industry norms for success and adopt and apply new models and metrics to prove it. Consumers have spoken very loudly about how they want to watch TV (wherever and whenever they want), and media owners are speaking clearly about how they will distribute content to make that possible.
Marketers must listen to them and collectively push for a new standard that defines, measures and attributes how to most effectively use TV as the great advertising platform that it is.