SAN JUAN, Puerto Rico — Brands relying exclusively on first-party data miss cross-brand consumer behavior patterns, creating blind spots that limit audience understanding beyond single-retailer interactions.
“Yes, that first party data is that liquid gold for a brand. A brand gets that from you because you want to then have your Kroger’s card or your Sephora points,” Margo Hock, vp for digital partners at Alliant, told Beet.TV contributor David Kaplan at the Beet Retreat San Juan. “So they get that information, but then they only know what I am doing there.”
Retail media networks and commerce partnerships capture behavior within individual brands but lack visibility into broader consumer spending patterns across categories and competitors.
Custom audiences require granular data
Custom audiences succeed in fragmented landscapes through tailored construction based on specific campaign objectives rather than off-the-shelf marketplace offerings, using granular data including income, family size, vehicle ownership, shopping patterns, and TV viewership.
“It’s not just going out into the DSPs, the marketplaces, and just getting all of these off the shelf kind of random things that media buyers are looking for,” Hock said. “If it’s not on the shelf, then what can we build behind the scenes? A lot of times that is that granular, raw data that we have behind the scenes that we don’t put out in the marketplaces.”
This enables consistent audience deployment across programmatic, connected TV, and social channels while providing unified measurement and ROI tracking.
Transaction plus behavioral data combination
Alliant’s acquisition of Analytics IQ last summer combined purchase-level transaction data with behavioral psychographic insights, creating differentiated capabilities beyond typical data partners.
“Alliant was known for that purchase level transaction data. And then Analytics IQ was known for that behavioral psychographic data,” Hock said. “We actually are combining two things that we don’t think a lot of other data partners have right now.”
The Analytics IQ backbone covers the entire U.S. population while transaction data represents approximately 90 million purchase records, enabling modeling that extends beyond direct transaction visibility to reach 265 million people.
Unknown data provides competitive advantage
Psychology-driven predictive insights focus on “not known” data points that competitors cannot easily access, particularly relevant during 2026’s significant political year when voting behaviors and related activities remain largely unmeasurable.
“We do have that deterministic data, but that’s kind of what all of our other competitors have as well — that known data like what kind of car you drive from the DMV or home loans and household income from the Census Bureau,” Hock said. “But then we go behind the scenes and we’re like, ‘What is this not known data?’“
AI adoption varies by vertical
Financial services clients require careful AI model construction due to regulatory constraints, while entertainment and CPG brands embrace optimization, proxy identification, and expanded targeting capabilities.
“We have to be very careful of how we’re building models, what AI is included in there and what we can and cannot do with that for them,” Hock said. “But then you have those entertainment, those CPG brands that are like, yes, we want that AI. We want to optimize. We want to find proxies.”
The industry continues grappling with fundamental questions about AI’s role in employment and operations.
“That seems to be the base of what we’re all talking about this week — is AI going to take our jobs or not,” Hock said.
You’re watching coverage from Beet Retreat San Juan 2026, presented by Alliant and TransUnion. For more videos from this series, please visit this page.
