"Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.
Today’s column is written by Eric Bosco, CEO at ChoiceStream.
Industry publications predict growing budgets dedicated to programmatic, which will consume larger percentages of brands’ total digital spending across various company sizes, verticals and countries.
In spite of this, there still exists ambiguity about how these types of media-buying systems work. That, coupled with the instantaneous nature of real-time bidding decisions and the anonymity of websites on which media buyers bid to serve, has caused an industrywide call for transparency.
There are many types of transparency that marketers and advertisers should seek on behalf of a brand, but one of the most important – audience insight – is often overlooked by industry pundits.
Transparency, when used in reference to behavior, can be defined as operating in a way that actions are visible or communicated. It’s characterized by visibility or accessibility of information, particularly regarding business practices.
Transparency in the context of a programmatic media-buying campaign could be referring to any of the following:
Price: How much a media-buying vendor pays for media vs. how much an advertiser must pay that vendor
Placement: The websites an ad is served on
Audience: The consumers who saw your ad
Audience insight: How audiences, defined by multiple segments, engaged with the brand, where and why
Audience insight is the most significant post-campaign or extended-campaign takeaway. Audience knowledge derived from algorithms should be shared with both the brand and its agency. Similarly, data visibility should go beyond basic delivery and performance metrics like number of impressions, clicks and dwell times. Programmatic partners should offer brands actionable information, including assets like the data-driven personas of the top-performing consumers, the days of the week and times of day during which peak performance occurs and the geography of top performers.
There are various layers of insight that a vendor can provide a brand, and each is strategically useful to the brand in efforts to connect with consumers. At a minimum, audience insight can be used in the design process of each campaign. It also is useful in identifying previously unknown audiences, influencing media buys in other channels like print and TV, and in guiding the development of partners and new offerings.
For example, if a customized phone cover manufacturer ran a campaign for discounted monogrammed iPhone covers, its programmatic partner might discover at the beginning of the campaign that females performed much better than males. The partner’s machine learning system would then begin testing other combinations with this gender segment, such as age, income or interests. It might learn that during the holiday season, mothers of a certain income engage the most with the ads, since they are buying gifts for their daughters, but during other times of the year, younger moms, teenage girls and women in their 20s are top intenders for the brand’s product.
Not only that, but the campaign may have experienced the highest conversion rates in suburban communities, and on Fridays, Sundays and Mondays from 8 a.m. to 10 a.m. and 7 p.m. to 10 p.m. Vendors should be open to describing this entire cycle to a brand because it could inform the brand’s messaging and positioning going forward.
Transparency is not for the machines; it’s for people. The data needs to be presented in a way that will tell a story to the people who read it.
Machine learning not only consumes big data, it also produces big data that can be unintelligible to most people. Efforts toward transparency mean reporting this data since it ties in to price metrics and campaign goals, but full transparency also means delivering personas that describe the traits and preferences of top consumers.
Whether a brand takes advantage of self-service or full-service programmatic media buying, performance information and audience insight should all be accessible throughout the campaign. It should be fit for consumption by those who will benefit from the insight it provides.