MediaMath is talking up its data management capabilities.
The company has rolled out a tool called Adaptive Segments, which lets a marketer gauge the performance of an audience and media placement before an ad runs.
The enhancement come on the heels of Salesforce’s $700 million agreement to buy data management platform Krux, an indication that DMP category is still white-hot three years after Neustar kicked off a wave of acquisitions by acquiring Aggregate Knowledge.
MediaMath is often thought of as a demand-side platform company, but it has data capabilities as well, including a DMP and a retail data co-op called Helix.
According to Keith Hawley, VP of audience platforms for MediaMath, the new tool’s aim is to fuel more intelligent bids in real time based on past purchase behavior or a user’s average lifetime value to a brand or marketer. In other words, it is aligning media activation and audience more closely.
“The most important impression you [identify] with Adaptive Segments is not always the impression you buy,” Hawley said. “You essentially eliminate a lot of impressions that aren’t valuable and you end up delivering more relevant messages to users that are seen.”
By getting more granular with audience segmentation, luxury Italian fashion brand Luisa Via Roma was able to generate up to 11.7 times higher return on ad spend using Adaptive Segments.
Although the idea of doing data segmentation based on known/CRM-based attributes (like who a company’s best, worst or lapsed customers are) is not new, what is new is the impact on media activation.
And there’s an efficiency angle as well, since advertisers can run “what if?” scenarios to test new segments rather than using their actual test budgets.
“I’m not only using (the DMP) for frequency capping, but to create coherent messaging that makes advertising more of a service than simply hitting you with a bunch of banners until you buy from me out of frustration,” said Véronique Franzen, programmatic marketing manager for Luisa Via Roma.
Because the brand caters to a diverse global customer base and sells products that range from $80 sneakers up to $18,000 high-fashion dresses, customer profiles vary greatly.
“What’s interesting for us is to cluster and regroup customers as we work through different marketing strategies,” Franzen said. “The fact that my insight into these audiences is updated in real time is very useful.”
Luisa Via Roma uses MediaMath’s DSP and DMP as a combo. Franzen claims doing so improves data consistency and basic usability.
MediaMath customers don’t need to use its combined DMP and DSP to access its Adaptive Segments. The data segments can be exported to a different DSP. But Hawley argues the combined approach reduces latency and data loss.
“We obviously think we have the most cohesive experience, but we don’t pretend the rest of the marketplace doesn’t exist,” he said.
“The DMP is a longer-term relationship,” he added. “Once you land on a DMP, it’s kind of like the difference between being married and dating. It’s where you flow all your data through, where you build APIs. We’ve had a DMP for a long time, but we’ve more recently invested in building up DMP products.”
