Discover
Inside the Stack
15 Episodes
Reverse
Consumer behavior has fundamentally shifted, and your next customer is likely making their purchase decision in a private group chat or DM . Yet many marketers are still relying on legacy attribution models that completely miss this private discovery . In this episode, host Lynne d Johnson is joined by Rick Lucas, Head of Performance Marketing GTM at Snap, and Ying Ying Kuang, CMO of Kitsch, to discuss how brands can evolve their measurement strategies . They explore Kitsch's recent geo-lift holdout test, the reality of living in the messy middle, and why Clicks + Deterministic Views (CDV) and causal testing are necessary to capture true demand . In this episode, we cover: The Shift to the DMs: Why authentic peer discovery and word-of-mouth recommendations are invisible to standard last-click platform reports . The Kitsch Geo-Lift Test: What Kitsch learned about incremental value after pausing Snapchat ads across roughly 16% of its markets for three weeks to measure performance against other channels . Clicks + Deterministic Views: How Snap champions CDV to capture view-driven conversions, and why last-click tracking can overestimate lower-funnel ad impact by up to 61% . The Messy Middle: Why incrementality testing is inherently choppy, why repurposing creative assets across platforms usually fails, and why winning requires zooming out to look at the full business P&L .
As CTV advertising becomes more sophisticated, local advertisers can target audiences with incredible precision. But there is a massive catch: the more precisely you target by audience, geography, and content, the smaller your available pool of households and impressions becomes. In this episode of AdExchanger's Inside the Stack, host Lynne d Johnson sits down with Adolfo Arbulú, Head of Pricing & Planning at Premion, to untangle the "Local CTV Paradox." Arbulú explains why stacking too many filters—like trying to target "hang-gliding nuns who drink energy drinks"—can accidentally shrink a local campaign down to just a handful of households. Tune in to hear why Premion built a proprietary identity graph to solve fragmentation, why you must decide on true measurement before a campaign launches, and why AI optimization in local markets still requires a "human in the lead" to navigate nuances like town events and political cycles. In this episode, we cover: The Local CTV Paradox: Why stacking too many audience filters suffocates a local campaign's reach. The Danger of Unmanaged Fragmentation: How chasing scale across multiple CTV partners without unified visibility leads to massive frequency overlap. Why Measurement Starts Upstream: Why proving business outcomes depends on the strategy and tracking established during initial planning, not just the dashboard at the end. AI and the "Human in the Lead": Why hourly real-time AI optimization creates noise in local campaigns, and why human strategists are required to navigate local market nuances.
Amazon Ads VP Kelly MacLean on how AI helps marketers connect the dots across the customer journey, with lower costs and better delivery in early tests.
AI is rapidly shaping how brands understand consumers, but what happens when you feed those models stale, fragmented data? On this episode of Inside the Stack, AdExchanger's Lynne d Johnson sits down with Matt Tipperreiter, VP of Product Strategy at Experian, and John Oronte, Director of Partnership Growth at AtData, to talk about the flaws hiding in modern consumer intelligence. Marketers have spent the last five years obsessed with hashed emails. But according to Oronte, a lot of those addresses aren't what they seem. When AtData looked at one client's verified new-to-file emails, almost 14% were burners set to expire within a month. On another platform, 15 to 20% of the records hadn't been used in six months. Brands are tying identity to stale emails, and AI is making decisions based on them. Tipperreiter and Oronte also get into data freshness across identity graphs, transparency, and whether you should own or rent your identity graph. And they explain identity to a five-year-old, with Tipperreiter turning his daughter's magnet tiles into the metaphor. This episode is brought to you by AtData, a part of Experian. AtData helps businesses unlock intelligence about the person behind the email address with email validation, data enrichment, identity matching, and fraud protection.
AI can build a marketing mix model in days, not months. But can you trust what it tells you? In this episode of AdExchanger's Inside the Stack, Madan Bharadwaj, founder and CEO of M-Squared, explains why faster modeling makes human judgment more important. With 20 years in the attribution space, Bharadwaj walks through why no single method is the truth and why causal attribution means triangulating MMM and incrementality testing. He also argues that a go-to-market thesis tells you whether a model is credible, and that a marketing accounting framework gives marketing and finance a shared language for outcomes like contribution margin. And then there's the story of how a beauty brand's AI-built model recommended cutting $30 million from its Amazon ad spend, and how SKU-level modeling led to a very different investment thesis and 5–7% top-line growth. Plus: why he thinks the future of measurement is super cognition.








