Zip Codes: The Simple Solution to Measuring Ad ROI

One of the hottest trends in ad effectiveness measurement, especially as privacy concerns kill user-level online tracking, is geo-incremental experiments. Done right, these experiments are cost-effective, straightforward, and reliable.

Geomedia experiments typically use large marketing areas, such as Nielsen’s Designated Market Areas (DMAs). Unlike traditional matched market tests, this modern approach involves randomly splitting DMAs, ideally all 210, into test and control groups. This way, advertisers with first-party data can measure real sales growth in-house without external services. For those without in-house sales data, third-party panels like Circana and NielsenIQ offer alternatives that are compatible with this type of test design.

High-quality, randomized controlled trials (RCTs)—similar to clinical trials in medicine—are the best source of evidence on cause-and-effect relationships, including the impact of advertising on sales.

Statistical models that include synthetic users, AI, machine learning, attribution, quasi-experiments of all kinds, and other observational methods are faster, more expensive, and less transparent forms of correlation—not causality. They may be effective for targeting audiences, but they are not for measuring ROI.

But imagine the potential of running geographic experiments using zip codes instead of DMAs.

Targeting by postal codes

One advantage of DMAs is that they are universally compatible with all media types. On the other hand, zip codes present challenges for experimentation in digital media. Targeting with online zip codes often relies on inferring IP addresses, which is unreliable and increasingly privacy-challenging. Geolocation signals from mobile devices also add zip codes to user profiles, which is bad for experimentation because a single device/account can be tagged with multiple zip codes based on where the user has recently visited.

The key to reliability in such geographic experiments is ensuring that the postal codes used for random media exposures match the postal codes from which viewers receive their bills, as recorded in company CRM databases. Each device and user should be targeted with only one postal code: the postal code in which they reside.

To adopt this technique, media companies need to take two transformative steps:

1. Use primary postcode targeting:
   Big players like Google and Meta already collect extensive user data, often tagging multiple zip codes on a single device. These companies should offer a “primary” zip code targeting option for experiments based on users’ profiles or the most frequently observed zip code for their devices.
2. Apply anonymous registration with postal codes:
   Broadcasters must register to access most free content and offer an “anonymous” account type that does not require an email address. Users provide a username, password, and home zip code, allowing broadcasters to develop audience profiles while maintaining user anonymity.