By Bridget Bidlack, SVP, Product Management, [x+1]
With a few years of programmatic media buying under their belts, marketers are starting to understand just how important it is to make their first-party data actionable. The early days of this media revolution were dominated by the unprecedented ability of DSPs to build a highly targeted audience at scale with look-a-like models using first and third-party data, and then to reach that audience across multiple inventory sources.
This was heady stuff back in 2010. Marketers were seeing some performance improvements and getting the reach that had eluded display campaigns pre-exchange buying. Ultimately, however, these audience segments were static and most DSPs could only do real-time optimization on contextual and technologic data attributes such as publisher, creative size and browser. With those limited capabilities, marketers aren’t much better off than doing a targeted network buy. Many DSPs didn’t (and still don’t) even have a flexible way of ingesting first-party data or have appropriate data standards that pass the security sniff test.
Optimizing campaign performance is the first step – but the real promise of programmatic buying is dynamically optimizing the actual audience segment. This can be accomplished by leveraging first-party data when it counts the most.
Consider this scenario: an audience of credit card intenders is built utilizing first- and third-party data for an activation campaign with a financial services brand. The campaign successfully generates traffic to a site where an offer is served to a visitor, compelling that visitor to fill out an application right that moment. Sounds like a home run, right? Not necessarily. Because the visitor did convert, they are no longer in-market for a credit card, but since the DSP doesn’t know this, it will mindlessly machine gun irrelevant ads at this poor person until the cookie expires. In this scenario the display channel and site channel operate in silos, neither considering what the other has or hasn’t accomplished.
How do we fix this? The solution is a more holistic strategy delivering seamless messaging that radiates out through multiple execution channels. Each interaction dynamically informs and predicts the next. Data, logic, and optimization are, quite literally, the center of the universe.
[x+1] Origin is a digital marketing hub. We have a DMP and a DSP that work together off the same cookie pool, meaning no data loss. We have a predictive optimization engine (POE) that optimizes in real-time using first- party, third-party and contextual data. We believe this should be a fundamental requirement for any DMP or DSP – but many other vendor solutions simply can’t do this.
POE puts marketers in control of their audience models and lends transparency into model output. See something you like? Save it and reuse it on your next campaign or even on your website. Have a hankering for bakeoffs? No problem, you can syndicate audiences found in POE models to another DSP. The difference with [x+1] audience syndication is that it’s better than pushing around a bunch of remarketing lists as there is real intelligence that goes into building the audiences. What about other channels? Leverage what POE has learned from your media campaigns and website optimization by syndicating those audiences to your email, SMS, call center and other vendors.
It’s simple, it’s easy and it works. Programmatic buying has come a long way since the early days of audience targeting, and savvy marketers are now unlocking the full power of digital marketing hubs that place first-party data at the center of their campaigns and marketing strategy.
Originally published by [x+Blog], [x+1] on August 15, 2013