Despite the many challenges brought on by the response to Covid-19, TV viewing is not only up, but so is wide consumer adoption of non-linear video platforms. And with that, ad units on PayTV AVOD, OTT and CTV are now a common line item on many brands’ media plans that traditionally have focused primarily on linear television. As this evolution scales, the work of manually gluing together linear and non-linear ad inventory to act uniformly is becoming more and more challenging for ad agencies and TV networks alike. As such, the march is on to streamline and unify linear and non-linear ad executions across platforms.
For years, PayTV AVOD, OTT, and CTV were considered video outliers that were “cool things” for the agencies to put on a brand’s media plan, and “gravy” revenue for the TV networks. Fast-forward to 2020 where up to 50% of a TV show’s viewing audience can be on non-linear platforms. “Cool” and “gravy” have been replaced by “firm delivery” and “monetization” against the brand’s KPIs. Unfortunately, obtaining firm delivery from various platforms can pose forecasting, delivery and measurement challenges. The good news is these execution challenges can be solved initially by managed, manual processes and automation in the ad tech-stack over time.
Let’s start with forecasting. An ad seller cannot accurately sell ad inventory until a forecast of that available inventory can be developed. Traditional linear systems are well-ordered and accurate at forecasting because the methodologies have been around since the dawn of television. When the TV universe, and past delivery of like-programming, are well understood, you can get a very accurate forecast.
The problem lies in the ability for today’s linear systems to quickly react, adjust execution, and reforecast delivery mid-flight based on fresh data. This linear limitation is accentuated when combined and compared with non-linear digital. One solution is to introduce dynamic ad insertion into traditional linear systems, an initiative that is finally starting to happen at scale. For non-linear systems, reacting to new data and adjusting accordingly is much easier, but an accurate forecast can still be challenging. This is mainly due to the fact that many non-linear systems do not yet fully understand the universe in which they operate, as they have been engineered solely for immediate ad decisioning.
The solution is to combine the time-tested methodologies of linear with the immediate ad decisioning of non-linear. Easier said than done, but it is possible, and being done today.
Next is delivery. You cannot effectively reach the ad campaign’s goals if you are not delivering the impressions as forecasted. While there are many delivery challenges that could be discussed, we’ll focus here on a major one, entertainment metadata. That is, consistently designating the title of a TV programme across both linear and non-linear platforms. Easy, right? Nope. I can’t go into why this happens in 600-800 words, but suffice it to say that if the programme title isn’t maintained consistently across platforms, it becomes very difficult, if not impossible, to accurately target an ad on that programme title consistently.
The solution has been to either attempt to ensure the same programme title metadata stay consistent from platform to platform from the top down, or to maintain a central programme title lookup table that the various platforms can reference from the bottom up. Both solutions have their challenges, but here again, both are being done today.
Finally, measurement. You cannot get paid for all this hard work if you cannot accurately report on what you have done. The first challenge involves a mismatch of what linear and non-linear platforms are measuring. Linear is based on a Gross Rating Point (GRP) and non-linear is based on an Impression. But here too, once you decide on what the universe is and what counts as an ad unit view, you can devise a methodology to combine the two. Perfect? Nope. But a usable approach for sure and is being done today.
And then there is Invalid Traffic. You cannot get paid if your customer is telling you your reporting in not valid. Linear and Pay TV AVOD have flourished simply by working from a single set of measurement “truth” in a closed loop ecosystem. But non-linear digital systems typically rely on multiple third parties who validate their truth in an open ecosystem. The problem is that sometimes these third parties interpret the valid behavior happening in the closed loop ecosystem as invalid. Correcting this requires constant, proactive communication with these third parties as to what is valid in the closed loop ecosystem so that the data can be “white labeled” as valid. Constant updating of the white label list is important, as it can change constantly.
So, as we march to streamline and unify linear and non-linear ad executions, let’s take faith in knowing these execution challenges can be solved by managed, manual processes – and in many cases, automation of the ad tech-stack is well underway.