The boundaries between Pay TV operators, OTT providers, and broadcasters are blurring. In the face of increased competition and cord-cutting, operators feel intensified pressure to diversify their revenues and launch new services. Targeted TV advertising is one approach operators can take to boost monetisation.
Targeted advertising has become a must-have strategy for every Pay TV operator and broadcaster. Netflix is the last streaming giant to recently announce that it will soon launch its own advertising tier. So, how can operators utilise targeted TV advertising in the best way possible? This article reviews strategic steps that operators can take to effectively implement AI/ML techniques in addressable TV advertising.
Key steps on the targeted TV advertising journey
Leveraging AI/ML for audience segmentation is a multi-phased process. The more accurate the segmentation, the smaller the audience for the targeted ad. By starting with simple use-cases and building up to granular segmentation, operators can gain more experience in targeted TV advertising, increase their knowledge of TV audiences, and ultimately, boost their revenues.
One of the first steps on the targeted TV advertising path is serving ads based on geolocation. While this may sound like a straightforward segment, where information can simply be pulled from the database, AI and ML are highly involved with achieving accurate and granular location-based targeting. Extracting accurate data from viewers who are watching content on the go, sharing credentials, and constantly moving between addresses is quite challenging without the proper AI/ML techniques in place. For this reason, the sheer ability of leveraging AI to target TV viewers based on geolocation is a huge game-changer, and geolocation addresses a little over half of the advertisers’ targeting ambitions, making it a must-have segment for operators. By utilising geolocation, operators can optimise the effectiveness of advertising and gain advertisers’ trust — as well as additional revenue opportunities.
The next tier of segmentation is usage-based. With this approach, operators can offer advertisers information about viewers who have already seen their ad on traditional broadcast TV — using their specific viewing history. This enables advertisers to either target or exclude certain individuals, based on their prior exposure to a certain campaign. ML allows operators to establish those segments with good granularity. Once the segments are in place, each ad can target a different, highly relevant audience.
Eventually, usage-based advertising enables operators to optimise their ads’ effectiveness and generate additional advertising opportunities by running multiple personalised commercials during a programme break. In addition, it also helps advertisers target “unreachable” audiences while they are watching VOD content.
Moreover, by using AI and ML, operators can deduce segments such as age, gender, and other demographic properties from the usage data. These segments are in high demand from advertisers. AI and ML allow operators to break down usage-based data and help advertisers deliver commercials oriented toward specific age groups or household structures. For example, operators might determine that the age group of a particular viewer is a teenager based on a viewing history of titles such as “Sponge Bob Square Pants” and “Saturday Night Live.” Operators can determine this by comparing the viewing history with a panel group and by applying ML techniques.
To extract more value from the data, and address insights that are notably based on changes in behavioral patterns over time, operators need to add a time axis. Advanced ML and AI techniques can identify pattern changes that point to life-altering events, such as starting primary school, retiring from work, going on maternity leave, and more. Eventually, gaining a deeper understanding of viewers and household life changes enables advertisers to deliver highly relevant ads.
Next steps: Putting the segmentation to work
Once operators have a well-established targeted TV advertising solution that can exploit the demand for mainstream segments, their next step on the targeted TV advertising journey is to put the segments to work. As the accuracy and granularity of the segments increases, the audience per segment becomes smaller. Sometimes video ads for highly granular audiences are too expensive to produce. In other words, operators might not gain the most out of the segmentation. A simple way to embrace granular segmentation is through banner ads and animated gifs. The cost of producing these is minimal and, combined with the advanced segmentation capabilities driven by AI and ML, they can allow for simple, quick, but highly effective advertising.
Additionally, Pay TV operators can leverage AI and ML to promote their own content. Statista found that ads are the most common way Americans and Canadians discover new movies and TV shows. Operators already have a large amount of data on usage consumption. The advanced capabilities of AI and ML algorithms can enable very accurate segmentation to effectively reach the desired audience.
Start simple and then expand
There is no better time than now to adopt targeted TV advertising. This burgeoning technology opens up new and expanded opportunities for Pay TV operators, allowing them to strengthen their position in today’s ultra-competitive TV market.
By starting with the simpler forms of AI and ML-based targeted TV advertising, such as geolocation and usage-driven ads, operators can quickly react to market demands. Then, operators may move on to a higher level of audience segmentation. By applying the latest AI and ML techniques to TV data, operators can deliver targeted TV ads that enable increased viewer engagement and create additional monetisation.
