Latest / The CMO Podcast with Fexingo: Marketing Leadership, Budgets, and Executive Strategy / Why CMOs Are Redistricting Media Budgets by Zip Code
Transcript
- Lucas: So you look at the media plan for a national brand, and it's still mostly built around DMAs — designated market areas — which are these huge tv shaped regions that haven't changed much since the 1990s. But increasingly, CMOs are asking: why am I buying ads for an entire metro area when 80 percent of my customers live in a handful of zip codes? Luna: Yeah, it feels like the pendulum is swinging hard toward hyperlocal. But is this actually new? Local advertising has been around forever — radio spots, newspaper inserts. Lucas: It is new in two ways. First, the granularity. We're not talking about city-level or even neighborhood-level. We're talking individual zip codes, sometimes even census-block groups. Second, the trigger. It's not just geography — it's geography fused with first-party purchase data and ai driven geomapping. Luna: Give me a concrete example. Who's actually doing this at scale? Lucas: Home Depot is a great case. They've been running something they internally call 'zip-first planning.' They take their loyalty card data — millions of transactions — and map every single purchase to the customer's home zip code. Then they build media plans that allocate spend proportionally to where their highest-value customers actually live. Luna: So instead of buying the Atlanta DMA — which covers like 30 counties — they're buying specific zip codes inside that DMA? Lucas: Exactly. And they're doing it across channels — connected TV, programmatic display, even out-of-home. The result: they've reported that zip code targeted campaigns see a 30 to 40 percent lower cost per visit than their broad DMA campaigns. Luna: That's a huge efficiency gain. But I have to imagine there's a privacy landmine here. When you combine zip code with purchase data, you're getting really close to identifying individuals. Lucas: It's a valid concern. The brands doing this well are very careful to operate in what they call 'aggregate anonymity.' They're not targeting individuals — they're targeting cohorts of at least 500 households. And they're using clean rooms and data clean rooms to make sure no individual-level data leaks into the bid stream. Luna: Still, the optics can be tricky. If a brand decides to pull spend from lower-income zip codes because the ROI isn't there, that could look like redlining. Lucas: That's the elephant in the room. And some CMOs are grappling with it head-on. I spoke with a grocery chain in the Midwest — they asked not to be named — and they reallocated 15 percent of their TV budget into zip code targeted programmatic ads served within a three-mile radius of their stores. Their ROI went up 22 percent. But they also realized they were effectively deprioritizing zip codes where their stores were still operating but where the average basket size was lower. Luna: So what did they do? Lucas: They created a separate budget line for 'access marketing' — a fixed allocation to maintain brand presence in every zip code where they have a store, regardless of immediate ROI. It's not an enormous amount, but it keeps them from creating marketing deserts. Luna: That's a thoughtful approach. It reminds me of the P&L structure we talked about a few episodes ago — where marketing has its own profit and loss. This feels like a natural extension, where you're not just optimizing for efficiency but also for equity. Lucas: Exactly. And the technology is only getting more precise. There's a company called Geobiquity — I think they're still private — that claims to predict foot traffic at the store level with 90 percent accuracy based on zip code level media exposure. If that holds up, it changes how you measure attribution entirely. Luna: It also changes creative. If you know someone in zip code 90210 versus 90011, you can serve them completely different ads, even within the same campaign. Lucas: Right. And some brands are doing that. Target, for example, has been testing zip code specific creative that highlights different products based on local purchasing patterns. A store in a coastal zip code might show beach gear, while a store in a mountain zip code shows snow gear — both from the same national campaign. Luna: That's smart. But doesn't it increase production costs significantly? Lucas: It can, but the brands that are doing it at scale are using dynamic creative optimization — DCO — where the ad is assembled programmatically based on signals. So you're not shooting 50 different commercials. You're shooting one modular commercial that swaps out scenes based on the zip code. Luna: That's the kind of efficiency that actually scales. So what's the downside? Why isn't every CMO doing this tomorrow? Lucas: A few reasons. First, the data infrastructure is expensive. You need a clean first-party data set, a data clean room, a DCO platform, and the talent to connect them. That's a multi million dollar investment. Second, the measurement standards aren't settled. There's no industry-wide definition of what 'zip code level ROI' means yet. Luna: And third? Lucas: Third, there's the agency compensation model. Most agencies still get paid on a percentage of media spend. If you shrink the spend by targeting only high roi zip codes, the agency's fee shrinks too. That creates a misalignment. Some CMOs are moving to fee for service or performance-based comp to fix it. Luna: So the shift to zip-code targeting isn't just a media decision — it's a structural decision that ripples through data, creative, measurement, and agency relationships. Lucas: Exactly. And that's why it's a cmo level conversation, not a media buyer level conversation. You're redesigning the entire go to market motion around a fundamentally more local view of the customer. Luna: You know, this kind of conversation is exactly why I love working on this podcast. It's practical, actionable, and it changes how you think about spending money. Lucas: I feel the same way. And if you've found today's episode useful, here's a thought: the reason we can have these deep, ad-free conversations is listener support. If you want to keep this going, you can buy us a coffee at buy me a coffee dot com slash fexingo. It's a small gesture that makes a big difference. Luna: Yeah, and every bit helps us stay independent and focused on what actually matters for marketers. Lucas: So back to the topic — I think the next frontier is going to be real-time zip code level bidding. Imagine a programmatic auction that factors in not just who you are, but exactly where you are, and adjusts the bid in milliseconds based on the store traffic patterns in that specific block. Luna: That sounds both incredibly efficient and a little dystopian. Are brands actually testing that? Lucas: A few are. There's a pilot happening right now with a convenience store chain in the Southeast that's using location data from mobile SDKs to trigger ads when a customer is within a quarter-mile of a store that's underperforming that hour. The creative changes: if it's 2 PM and the store is quiet, the ad says 'afternoon break, 20 percent off any coffee.' Luna: So it's demand-response marketing at the hyperlocal level. That's fascinating. Lucas: It is. And it raises a question I think every CMO needs to ask themselves: if you could know exactly which zip codes drive your business, and you could tailor media and creative to those zip codes in real time, would you? And if not, why? Luna: That's the kind of question that separates leaders who are optimizing from those who are transforming. Great episode. Lucas: Thanks for listening — we'll be back next week with another conversation that digs into the real numbers behind marketing decisions.