As noted in last week’s Advanced TV Insider, a new report lays out the progress and remaining hurdles for free, ad-supported streaming services (FASTs).
The report, authored by TVREV’s Alan Wolk and Mike Shields, also postulates that contextual targeting might be the solution to many of this new media platform's challenges.
Given that FASTs are clearly becoming one of the fastest-growing channels for advertising investment, it seems worth summarizing that argument here, for readers’ evaluation.
First, they point out that advertisers — particularly those with large budgets — do want to be able to target by content, as they can with traditional TV, because it’s been demonstrated to improve performance. For instance, according to Vizio, which uses ACR data in part to target users based on what they watch and have watched, including preferred content genres, campaign data shows that ads do tend to work better when there is strong contextual alignment.
Here's how contextual targeting might help with FAST’s most common drawbacks:
Reach: Advertisers are frustrated by their inability to buy across multiple FASTs, as they can with TV networks. The core problem lies not so much in technology, but the fact that FASTs have created their own linear channels that each have their own unique content. That precludes the kind of buy possible with, for example, NBCU’s SyFy network, which allows for targeting science fiction fans across hundreds of different multiplatform video programming distributors (MVPDs).
Allowing for a contextual buy against “science fiction content” would allow an advertiser to reach a sizable audience across multiple FASTs and SVODs — not all that dissimilar to using keywords to buy segments in digital media.
In addition, advertisers would be able to use KPIs that show they’ve reached everyone who is watching the TV in a home, not just the one person in their target demo — “thus making it much easier to achieve the sort of massive reach that big brands are looking for when they add TV to their media mix,” notes the report.
Privacy: As advertisers and programmers have come to realize, using individual identifiers to target within streamed TV poses thorny privacy issues, including: Have viewers given permission to have their personal data collected and used for targeting, and if so, to and by whom? Is the device ID, or an IP address, being used to target? And with most viewing being household viewing, how is the actual "viewer" being determined?
“Contextual targeting removes many privacy issues by targeting households based on what is on the glass, not by who is watching,” says the report. “This allows advertisers to reach a broader audience and to avoid all of the issues involved in trying to activate digital-style targeting on TV.”
Transparency issues: When advertisers seek to target individuals, there are privacy issues around device IDs and IP addresses, cherry-picking issues around targeting users based on specific shows, and technology issues, because programmatic advertising is designed to seek out a specific viewer, not a specific genre.
“Contextual may be the solution, as it is easy to identify the shows based on the tags and universal content IDs that allow them to be classified by genre,” the report points out. “There’s also less of an issue around cherry-picking if the goal is to reach viewers against specific moments during a show.”
ID resolution and datasets: With contextual targeting, brands don’t have to worry about collecting or paying for often-inaccurate first-party data, or targeting too narrow a group. They can pick where their ads run based on what the show is about, who is likely to watch it and how it fits with the message and tone of their commercials.
Measurement: While contextual advertising may not be able to solve all of streaming’s measurement issues, something like the IRIS_ID — a universal content ID that allows content to be sorted into context-based datasets — might provide some relief, say the authors.
Reason: The IRIS_ID makes it easy to identify programs without having to rely on device IDs or IP addresses and provides a consistent nomenclature for shows, as well as data around the content across ACR providers, who do not pool their data. Importantly, it also makes it easier for advertisers to use their measurement provider of choice. The more common contextual targeting becomes, the easier it will be to identify shows and which ads ran on which shows on which platforms.
Excessive frequency: “Contextual targeting’s ability to ameliorate the issue of the same households getting served the same ads over and over again will rely on how tightly the contextual target is defined,” the report acknowledges. “But by focusing on content types rather than on specific viewers, it will likely provide a much broader pool of programming on which ads can run, and that, in and of itself, can make it less likely that the same ad is served up multiple times within the same pod.”
The report stresses that the advantages of contextual targeting in no way mean that demographic targeting would cease to play an important role in streaming.
While contextual is well-suited for big brands that mostly run image ads and see TV as a reach vehicle, demo-based targeting is what has opened up streaming TV advertising to performance-driven D2C brands. “Similarly, for brands who see the FASTs as a reach extension vehicle, a way to get their message in front of viewers they are missing on linear and (soon enough) the SVOD services, demo-based targeting makes it easier for them to ensure they have met their reach goals,” the report states.
Further, demographic and contextual targeting can be used together effectively — for instance, by a brand seeking to target viewers of a particular content genre within a particular DMA — a use case likely to become more common as FAST advertising matures.
So what’s the main holdup in putting contextual targeting capabilities into place across FASTs and SVODs?
Technology? Nope. It’s that old standardization nemesis nomenclature.
The industry hasn’t agreed on a common metadata set that would enable easy, efficient buys across the big vendors.
“We still don’t have enough of a common nomenclature across every platform, especially on higher-end content, to pull off a big-budget buy across multiple FAST and SVOD services,” notes a senior ad agency executive.While IRIS.TV and other companies are working on this, “until programmers are convinced that creating this common metadata solution is in their best interests, it will be a challenge to get big brands to make big budget buys,” concludes the report, adding that education is the key to getting through to programmers.
There is no reason to believe that if a TV commercial appears in a break in a rustic sitcom that it will communicate its sales pitch or motivate those viewing to a greater degree than if the same commercial reaching the same kinds of people was seen in a detective show or a medical drama or a reality talent show or a daytime talk show. What can happen is that a drama will tend to hold more of its viewers in the room during a break as they are more engrossed by its content and don't want to miss the continuation of the dramatic content. So you get somewhat higher ad attentiveness---eyes-on-screen ----factors in dramas and somewhat less in other types of shows. This does not autoimatically mean that those viewing buy your claims.
What TV advertisers probably mean when they use the term "context" is nothing more than their notion of demographic targeting---something they have been doing---- and not always brilliantly---since the mid-1960s. If you are targeting young, you try to orient your time buy towards those shows --"contexts"--that have high concentrations of younger viewers. If you are going upscale you try to find shows that reach reasonable percentages of affluent viewers, etc.
In addition, there are certain "contexts" that are so firmly built into the advertiser's consciousness---or promotional strategy that they have become must buys. These are news, sports and big event specials. A finacial services company is going to be in news no matter how old its demos are and no matter how high its CPMs. A sports car or beer brand is going to buy into the NFL games no matter what the price, etc. As somewhere between 37-45% of' "linear TV" ad dollars goes to these three "contexts"---news, sports and specials---it's not surprising that such advertisers think that "contextual" buying is the way to go. They think they are already doing it. Many TV advertisers who use other program genres agree with them.