Where Your Ads Stop Working: Finding the Dead Zones
Here’s something your media vendors will never tell you: your ads are running in geographies where nobody’s responding.
Not markets where you’re not buying. Markets where you are buying, the impressions are being delivered — broadcast, radio, programmatic, CTV, outdoor — and the digital response is effectively zero. No web visits. No searches. No phone calls. Nothing.
I call these Dead Zones. They exist in every campaign I’ve ever analyzed — across every medium with geographic delivery. And they’re invisible unless you measure at the zip code level within each market.
What a Dead Zone looks like
Let me make this concrete. A national home services advertiser runs linear TV across 40 markets. In the Dallas-Ft. Worth DMA, the campaign is running across four network affiliates and two cable networks. Total coverage is strong. Ratings are solid.
But when you look at the zip-level response data, something emerges. The northern suburbs — Plano, Frisco, Richardson — show strong digital engagement. Web sessions spike after ad airings. Phone calls come in. The pattern matches what you’d expect from an effective campaign.
The southern suburbs — a cluster of zip codes in the opposite direction — show almost nothing. Population density is comparable. Demographics are similar. Household income is in the same range. The ads are airing on the same stations, in the same dayparts, with the same creative. But the digital response is near zero.
That’s a Dead Zone. The ad is reaching the market. The market isn’t responding. And without zip-level geographic data, you’d never know — because the Dallas DMA average looks fine. The high-performing north is masking the non-performing south.
Why Dead Zones exist
Dead Zones aren’t random. They have causes, and understanding them changes what you do about them.
Coverage and competition gaps
Cable zone boundaries. Within a single DMA, cable operators divide coverage into zones. A cable ad buy might cover one zone but not another. Linear TV targeting is primarily DMA-based — every household watching a show in a region sees the same broadcast ad. But cable buys are more granular. If your cable weight is concentrated in certain zones, the zip codes outside those zones are getting broadcast-only coverage, which may not be enough to drive response.
Competitive saturation. In some zip codes, a competitor’s advertising weight overwhelms yours. The viewer sees your ad but also sees three competitors’ ads in the same break. The response scatters. Your ad is one signal in a noisy environment. In neighboring zip codes where you have less competition, the same creative produces stronger response because it has more relative weight.
Audience and viewing differences
Audience mismatch. The DMA-level demographic data says the market fits your target. But zip-level demographics tell a different story. A zip code within the DMA might skew younger, older, lower-income, or toward renters rather than homeowners — any of which could explain why a home services ad isn’t producing response.
Content context. The same ad on the same station at the same time can drive different response depending on what else is on in that market. Local programming, local news anchors, local competitive activity — these vary by geography in ways that DMA-level measurement can’t capture.
Cord-cutting patterns. Streaming surpassed linear TV viewership in 2025 — 44.8% for streaming vs. 44.2% for linear. But cord-cutting isn’t uniform geographically. Some zip codes have higher cord-cutting rates than others. If a zip code has 60% cord-cutting penetration, your linear TV ad is reaching a smaller fraction of the population than the DMA average suggests. The ad is airing. Fewer people are watching.
Dead Zones aren’t evidence that TV doesn’t work. They’re evidence that TV works unevenly — and that the unevenness is invisible without zip-level geographic data.
Why you can’t see them without zip-level data
Most TV measurement operates at the DMA level. There are 210 US DMAs, each containing dozens to hundreds of zip codes. The Dallas-Ft. Worth DMA, for example, has over 500 zip codes. Measuring at the DMA level means treating those 500+ zip codes as one unit.
When a DMA shows 15% lift, it doesn’t mean every zip code within it has 15% lift. It means the average across 500+ zip codes is 15%. Some zip codes might be at 40%. Others at zero. The average tells you nothing about the distribution.
This is the resolution problem. Linear TV ad spending is projected to decline to its lowest share since 2005 — marketers are reallocating 36% of linear TV spend to CTV because CTV can target at the zip code level or even the household level. Linear TV is losing the measurement argument because it measures at a resolution that hides its own inefficiency.
The irony is that linear TV’s delivery is geographic. Broadcast signals have physical boundaries. A local affiliate covers a specific polygon — not an approximation, a real delivery footprint. The geographic data exists. Most measurement platforms just don’t use it at the resolution needed to find the Dead Zones.
How to find them
Finding Dead Zones requires three layers of geographic data, compared simultaneously.
Three layers of geographic data
Layer 1: Delivery footprint. Where is each ad actually airing? Not “the Dallas DMA” — which specific publishers are delivering, and what are their actual coverage polygons? A local NBC affiliate has a different coverage area than a local FOX affiliate. A national cable network covers the entire DMA but with different cable zone penetration.
Layer 2: Response geography. Where is the response coming from? Every web session resolves to a zip code through IP geolocation. Every phone call resolves through caller ID and geographic matching. Every CRM record resolves through the customer’s service address. Response geography is specific — not DMA-level, zip-level.
Layer 3: Baseline comparison. What’s the expected response in each zip code? Population density, household count, historical response patterns, demographic profile — all of these set the baseline for what “normal” looks like in that geography. A zip code with 50,000 households should produce proportionally more response than one with 5,000. When it doesn’t, that’s a flag.
The Dead Zone identification is the gap: zip codes where delivery is confirmed (the ad is airing), population and demographics are comparable to responding zip codes, but digital response is significantly below expected levels.
The scale of the problem
Let me give you a sense of the scale. A single TV market — say, Dallas-Ft. Worth — contains over 500 zip codes. Across 40 markets in a national campaign, that’s 10,000+ geographic units to analyze. Each one needs its delivery footprint mapped, its response measured, and its baseline established. Multiply by the number of publishers, dayparts, and creative executions, and the computational requirement is enormous. This isn’t something you do in a spreadsheet. It requires infrastructure built specifically for geographic analysis at this resolution — a unified geographic entity layer with over a million entries, covering every zip code, city, metro, county, and DMA across North America.
That infrastructure is why most measurement platforms don’t offer zip-level analysis. It’s not that the math is hard. It’s that the geographic data layer doesn’t exist in most systems. They have DMAs. They don’t have zip-level delivery footprints.
This is what the Insights & Data Engine does across all 254 TV markets in North America. Every publisher’s delivery footprint mapped as actual polygons. Every response resolved to its geographic origin at the zip level. Every zip code compared against its expected baseline. The Dead Zones emerge from the data — they’re not modeled or estimated, they’re observed.
What to do about Dead Zones
Finding them is only useful if you act on them. Here’s where the geography pillar turns into a media strategy.
Fill gaps with digital
Supplement with programmatic. Dead Zone zip codes become targeting lists for programmatic display and video. The audience is there — they live in the DMA, they match the demographic profile. They’re just not responding to linear TV. A programmatic campaign targeted specifically to those zip codes fills the gap that linear is missing. CTV platforms can target at the zip code level — put CTV impressions in the exact geographies where linear isn’t working.
Adjust linear weight. If certain cable zones within a DMA consistently produce Dead Zones, consider shifting cable weight to zones that respond. This requires working with your media buyer, but the data makes the conversation specific: “these 15 zip codes in the southern suburbs aren’t responding to cable — shift that weight to the northern zones where we’re seeing 30% lift.”
Test creative by geography. A Dead Zone might not be a media problem — it might be a creative problem. If the same creative produces strong response in some zip codes and nothing in others with similar demographics, the message might not resonate with a geographic subculture or local market condition. Testing alternative creative in Dead Zone geographies can reveal whether the issue is delivery or message.
Close the loop on performance
Feed it back to your platforms. When you upload Dead Zone data to Google Ads as geographic bid modifiers, search campaigns adjust automatically. Bid more aggressively in responding zip codes where TV is creating demand. Bid more conservatively — or supplement with different creative — in Dead Zone areas where TV isn’t producing the search intent your paid campaigns need.
Measure the fill-in. When you deploy programmatic or CTV to supplement Dead Zones, the measurement system traces the response through the same geographic stack. Did the fill-in campaign produce the engagement that linear missed? The data shows it — or shows that the problem isn’t media at all, but something structural about those geographies.
Build proprietary audiences
Build audiences from the gaps. Dead Zone zip codes aren’t just targeting lists — they’re the foundation for proprietary audience segments. The people who live in those zip codes, match the demographic profile, and aren’t responding to your media become a targetable audience across programmatic, CTV, and paid search. These are audiences that no DSP carries and no data marketplace sells, because they can only be built from the intersection of your specific delivery data and your specific response data. An advertiser in a different category would have completely different Dead Zones. The audiences are unique to your campaign, your markets, and your data.
Dead Zones are a feature, not a bug
Here’s the reframe I want advertisers to consider: Dead Zones aren’t evidence of failure. They’re evidence of precision.
Seeing variation is the advantage
Every campaign has geographic variation. Every media plan has markets that over-perform and markets that under-perform. The difference is whether you can see it. National/local TV ad silos are collapsing as the industry moves toward converged measurement. The advertisers who can see their geographic performance at the zip level make fundamentally different decisions than the ones looking at DMA averages.
A DMA average that says “the media is working” is a reason to renew the buy. A zip-level map that shows exactly where it’s working and where it isn’t is a reason to optimize the buy. One keeps the budget. The other improves the ROI.
Resolution determines decisions
The IAB’s 2026 State of Data report found that 75% of buy-side leaders say measurement approaches are underperforming. Part of that underperformance is the resolution. DMA-level measurement produces DMA-level decisions. Zip-level measurement produces zip-level optimization. The data exists. The infrastructure to analyze it exists. The question is whether your measurement vendor operates at the resolution that reveals the Dead Zones — or the resolution that hides them.
Every campaign I’ve analyzed has Dead Zones. The only variable is whether the advertiser knows about them.
Here’s the question to ask your current measurement vendor: can you show me the zip codes within each DMA where my ads are airing but digital response is below expected levels? If the answer is no — if they can only show you DMA-level averages — you’re flying blind on the geographic dimension that determines whether your spend is working or wasting.
The Dead Zones are there. They’re in your data right now. The only question is whether anyone’s looking at the right resolution to find them.