A Solution for the CTV Content Identification Problem

Programmatic is favored by the buy side for its speed and ease and built-in targeting and data feedback options. Agencies love that programmatic reduces labor costs, making media spend more profitable for the agency. (Although programmatic can be a leveler further commoditizing media agencies unless they can find a way to prove they have a secret sauce approach to programmatic and can prove that it works in terms of measured business outcomes.)

Yet despite the love affair with programmatic CTV, there are a couple of flies in the ointment.

One of them that is well known is fraud. Unless one sticks with the major television networks, there is a significant chance that the impression will wind up behind the frame that is viewable on the screen or on content that would not please the client. This goes beyond fraud to brand safety concerns. There are solutions for these problems, including the MRC viewability standard and services which help filter out these unwelcome environments. Paying the viewable CPM, however, goes against the ingrained instincts of the buy side which has traditionally been steeped in making CPM the most important catechism, and the same goes for paying for the filtration services. This fly in the ointment is today under a degree of relative control and so is not the center of our focus here.

The other anointed fly is that you don’t really know what you’re buying. This limitation was first laid down in the streaming world by the pioneer Hulu. Their position reflected, to a large extent, the existence of a US federal law called VPPA, the Video Privacy Protection Act of 1988. The VPPA was enacted when a Supreme Court nominee was embarrassed to have been caught by the press for having rented porn videos. The law specified video tapes but in court cases in 2024, the courts supported the interpretation that this should also cover the likes of pixel-based tracking in CTV.

The way that streamers today look at the situation is that they cannot provide to the programmatic bidder (or any other buyer) both the ID and the content identification; they can present one or the other but not both. From a legal point of view, this is really a matter of interpretation, because the IDs used in selling ID-based audience are supposed to be non-PII, that is, not reidentifiable. However, different States have their own privacy rules and the risks of privacy invasion are properly perceived as enormous by marketers and by media alike. Not enormous in terms of probabilities, but in terms of the amount of damage which could result. This is what causes the industry to hold back on providing both a program identifier and a non-PII ID.

Another factor is that a non-PII ID can be reidentified by someone expert in the art, or someone who has a large first-party database, or access to a state-of-the-art ID graph.

So here we are, CTV a wonderful medium, especially the longform off-air segment (often called OTT) shown by Bill Harvey Consulting research sponsored by FOX to be even more ROI impactful than linear TV, and far more ROI impactful than online shortform video (the other part of CTV), which is itself more ROI impactful than digital display… but hampered by the inability to tell the buyer the name of the program she or he is able to buy or bid upon.

Magnite and PubMatic will tell you the network that the program is on, which is a help, and more CTV platforms should be doing that. It is also possible to buy directly from networks, but this is more work as it is not programmatic and involves many deals instead of one deal. It also tends to be more expensive in terms of CPM, although this is not always the case.

Here is one more proposed solution that is inexpensive, easy, and available immediately. It avoids revealing the name of the program, but what it substitutes is actually something more valuable than the name of the program. Instead of the name, a new metric which already exists is provided, like an attention score, but it is more than that, it is a validated predicted sales effect score.

If you know the name of the program and are familiar with that program you can bid with confidence that the client will have no quarrel with the buy, except if you know it is blacklisted by the client. If you know the predicted sales effect score, it’s even better than that, you can bid more if you see a high predicted sales effect score, or bid lower if the score is lower. In other words, knowing the name of the content is not an end in itself, it is a means to an end of capturing the content that will do the most good for the client, and having a score is better than having a name.

The score of which we speak is the much-validated RMT Resonance score, which is a measure of the degree of congruence between the motivations in the client’s ad and in the specific program. RMT is a company I co-founded, and I am Chairman of RMT. The whole reason I developed the RMT method is to increase advertising effectiveness per dollar; the same reason I developed the ADI, set top box data, addressable commercials, TRA, etc. At the time, I did not know that things would come to pass that would lead to an impasse in the full flowering of OTT/CTV and that RMT would hold the answer to the problem, but here we are.

Posted at MediaVillage through the Thought Leadership self-publishing platform.

Click the social buttons to share this story with colleagues and friends.
The opinions expressed here are the author's views and do not necessarily represent the views of MediaVillage.org/MyersBizNet.