Commentary
TV Listings Data Will Be King
- by Michael Kokernak , November 11, 2009
Most television media plans today are partially built using fusion data gleaned from consumer research. Even with all the indicators of an interactive future, TV media planning has not made
the leap to its metadata future. Realistically, interactive television planning will also have to work with incumbent media software systems. Thus the granularity of listings data (also known as
metadata) is sure to grow in importance.
With video content delivery further decomposing into linear, time-shifting Internet, and video-on-demand buckets growing, it will be almost
impossible for planners to hit their consumer targets. Fusion with consumer panels, while accepted today, will most likely not be the best way to build flights, as planners need something to
help them navigate all the new viewer data.
The challenge then is to build a data scheme within today's media-buying software that can help planners make sense of all of
television. The data scheme would need to be able to tie together all forms of video content, so planning ccould be coordinated and results measured in a timely fashion. I believe that TV
listings data is that universal scheme -- and interactive TV "clicks" could become the primary data source for planning.
The infrastructure for this data scheme exists today
within the television listings companies. Each day broadcasters and cable networks update their schedules to centralized third-party databases that then syndicate out the listings to mobile
carriers, electronic program guides, Internet guides, media software, to name a few. Today listings information is nearly error-free, robust, and can be updated in near real time.
Planners will need to sort schedules using perimeters like program name, genre, time of day, actor, and Zip codes when trying to determine the consumer "engagement" baseline for their
campaigns. Using this method, planners should then be able to buy flights queried from TV listings attributes, historical click data, and ratings. Fusion research from consumer panels, in my
opinion, would become more of a forward-looking planning tool, only for those campaigns that had no historical TV click data.
Listings data will also open up the long tail of television
advertising so that companies may be able to target a very narrow audience (like rodeo shows). We may find that as time goes on, advertisers should be able to buy individual interactive avails, or
lower-thirds, similar to how keywords are sold on the Internet (by show, actor, genre, location, etc). In addition, listings data will be able to bridge the gap and link noninteractive avails to
interactive avails, so that the transition to nationwide interactivity campaign analysis is not patchwork.
As always Michael is either on the horizon or over the other side. One of these days, forces will be joined in today's world and his dream will finally be fulfilled.
Your assertion is a good one, however, there are many obstacles implementing it. A few key ones are:
(1) a program's listings aren't necessary consistent across markets. Stations submit their schedules individually and, other than character restrictions, have leeway in how they refer to the program. This is more of a syndication issue but it can apply more broadly.
(2) Not every program has episode detail associated with it. Ergo, it may be difficult to connect a buy at a level below program.
(3) On Demand programs aren't listed through the central listing services. Therefore, other than through Nielsen, I'm sure how you can connect these assets across platforms.
(4) Broadband sites may use a universal nomenclature--somewhat dictated by the content owner--to identify programs and episodes, however, you would also have to link the ads across these sites and potentially ad networks in order to generate a composited view of the asset's audience.
(5) A spot that is inserted into a program would have to be displayed across all platforms in order to add the audience together. Not hard to implement, but may be hard to identify and count, i.e., different creative for the same product by platform, region, etc.
Lastly, even if these and other barriers are overcome, I still question whether Planners really want to take the time and effort to do so -- is this a real goal for Planners or is it just posturing for the sake of their clients (the advertisers)? In other words, how close is close enough in an all digital world where too much data is as frustrating as too little data. If advertisers really want one-to-one relationship with consumers in order to provide better goods and services, they should seriously consider new methods and models for engendering or for fortifying those relationships rather than trying to retrofit existing media measurement and models.