Brands know their audiences; there’s no contesting that. The services and products that they sell are designed in the very first place to meet the needs of a core customer group that is well defined by a few common attributes. While the most digital advertising campaigns often involve a very pointed grouping within an audience that shows the most engagement and achieves the most conversions, there are also prospects that exist outside of the defined core.
A software brand, for instance, spends its ad dollars to reach the demographic that it knows from historical sales is likely to purchase its software: VP-level executives in their 40s and 50s. The brand sets up a set-it-and-forget-it campaign to reach these consumers. This is a partially wasted opportunity. While a majority of buyers may match this persona, many people who match the persona will never have anything to do with software decision-making. In this case, it is actually a younger, mid-level manager who is the true decision-maker about buying software; this person’s boss remains uninvolved. The brand cannot discover this with a campaign that has pre-set targeting; it sets up a strict pool of whom to target, and lacks the ability to discover and optimize to top-performing audiences.
By running a programmatic media-buying campaign, the brand marketers for the software company can further analyze the complexities of pre-existing customers, thereby avoiding much of the waste, while discovering new audiences. Machine learning also allows a system to identify which of those consumers being targeted are actually converting. For example, VPs in their 40s with no children and VPs in their 50s with children above 20 years old may be more likely to convert, because they have more time to take on the task of vetting software providers.
In a nine-month performance-based programmatic campaign for a national banking company, data from the campaign informed the brand about the top-performing persona’s lifestyles, behaviors and motivations. Campaign data informed the brand that its highest-performing persona was a young woman with a low household income who was starting her first job, renting, and/or beginning financial planning. By identifying the intricacies of the top-performer and then targeting this persona at scale, the campaign exceeded conversion goals. The brand achieved an average of 20 checking account application starts and two application-completes per day, while also securing a high of 98 application starts in a three-day period.
Simplistic targeting, or choosing a single segment based on gender, age, wealth, ethnicity and other attributes alone, will not bring a brand the success that it can expect from more sophisticated methods. Engagement and performance with branding and direct response campaigns thrive best when the campaigns are run through programmatic platforms with algorithms that optimize audience targeting. Providing a system with initial targeting guidance rather than strict guidelines allows a machine to discover on its own, in real-time, the top performing audience at each time of day and day of week during a campaign. Often times, one persona performs best during the workday, but another does well at night or on a weekend. Automated systems discover this and adjust targeting accordingly as audiences change.