What Is Geopath, and How Is an OOH Rating Actually Built?

Every OOH proposal you receive quotes impressions. Most of them, in the United States, trace back to Geopath. Very few buyers can explain how that number was constructed — which means very few can tell when it is being used well and when it is being used as decoration. This post walks the chain end to end.

What is Geopath?

Geopath is the out-of-home industry's audience measurement organisation in the United States — the OOH equivalent of what Nielsen is to television. It maintains a database of hundreds of thousands of media units and publishes standardised audience estimates for them, so that a billboard in Dallas and a transit shelter in Boston can be compared on the same basis.

It is a non-profit tripartite body: media owners, agencies and advertisers all sit inside it. That structure matters, because it is the reason a Geopath number carries more weight than a media owner's internal estimate — the people being measured do not control the methodology alone. Geopath's own research methodology documentation is public.

How is a Geopath OOH rating constructed?

Geopath builds a rating in a chain of steps, each one narrowing a large, crude number into a smaller, more defensible one. Per Geopath's Constructing An Out-Of-Home Rating, the sequence runs roughly like this:

1. Circulation. Total traffic volume past the display, derived from trip movement data across hundreds of millions of mobile devices and connected cars, converted into people passing during the week.
2. Demographics. Trips are connected back to census-block-level residence, producing anonymised demographic profiles of who passes — not just how many.
3. Geo-positioning. The display is mapped precisely relative to the roadway it draws audience from, capturing the factors that affect whether the ad can be seen at all.
4. Contact zone. Road type and display size determine the maximum noting distance — the area from which the ad is legible.
5. Speed data. INRIX traffic data sets travel speed past the display by hour and by day, and weekly circulation is allocated across those hours.
6. Dwell time. How long people are inside the contact zone, hour by hour. Longer dwell, higher likelihood of noting.
7. Standard ad noting. Eye-tracking research strips out people who had the opportunity to see but did not actually notice, adjusted for size, positioning, noting distance and speed of passage.
8. Number of digital ads. For digital units, spot length and rotation size are recorded — an eight-second spot in an eight-ad loop, for example.
9. Opportunity to see a digital ad. Dwell time and rotation are combined: in moderate congestion, a passer-by might have the opportunity to see three of the eight ads in the loop.
10. Digital spot noting. Eye-tracking applied at the spot level, producing the number of people who see each individual digital spot.

The important thing to notice is direction. Every step after circulation makes the number smaller. That is what a measurement standard is for.

What is the difference between OTS and a Geopath impression?

Opportunity to see (OTS) counts everyone who passed within range. A Geopath impression counts those estimated to have actually noticed the ad, after the visibility adjustment. The gap between the two is the entire value of the methodology.

This is why a "3 million impressions" claim means nothing without its basis. Three million OTS and three million visibility-adjusted impressions are not the same product, and the second one is materially more expensive per unit for a good reason. We took this apart in more depth in Plays vs. Impressions: What You're Actually Buying in DOOH.

What are OOH ratings, GRPs and reach?

A rating is the audience of a display expressed as a percentage of a defined market population. Sum the ratings across your buy and you get gross rating points (GRPs) — the standard currency for comparing OOH weight across markets of different sizes.

From the same underlying data you can derive reach (how many distinct people saw the campaign at least once) and frequency (how many times, on average, each of them did). GRPs are reach multiplied by frequency, which is why two campaigns with identical GRPs can behave completely differently: one broad and shallow, one narrow and repetitive. If a proposal quotes GRPs without the reach/frequency split, ask for the split.

What does Geopath not tell you?

Geopath measures exposure. It does not measure whether anyone did anything.

It will not tell you that a passer-by visited your store, searched your brand, or bought anything. Attribution — connecting exposure to outcome — is a separate discipline layered on top, usually via location-based visit lift, geo-holdout testing, or matching exposure windows against your own conversion data. We laid out that framework in How to Measure Taxi Advertising ROI.

Confusing the two is the most common mistake in OOH reporting. A perfect Geopath number tells you the ad had a real audience. It says nothing about whether the campaign worked.

What should you ask a media owner about their Geopath numbers?

Four questions do most of the work:

• Are these units actually Geopath-measured, or estimated by analogy? Not every inventory unit in a proposal carries its own measurement. Some are modelled from "comparable" units. That is a legitimate practice and a legitimate thing to disclose.
• Are the impressions visibility-adjusted or raw OTS? See above. This single distinction can move a number by a large multiple.
• What audience definition is this? Geopath reports A18+ by default, but demographic targets produce much smaller numbers. Make sure you are not comparing an A18+ total against a competitor's target-demo total.
• What is the reporting period and share of loop? For digital, impressions scale directly with your share of the rotation. A number without a stated share of voice is unanchored.

How should Geopath data sit on a campaign dashboard?

Planned versus delivered, per unit, per day — with the Geopath planned estimate as one line and actual delivery as another. The interesting signal is never the total; it is the drift between what was sold and what ran.

Keep the modelled layer (Geopath estimates) visually distinct from the counted layer (plays, screen uptime, verified delivery) and from the outcome layer (visits, searches, conversions). Buyers lose trust in OOH reporting when those three are blended into one confident number. They keep trust when each is labelled for what it is.

FAQ

Is Geopath used outside the United States? No. Geopath covers the US. The UK equivalent is Route; other markets have their own bodies. Do not compare a Route number to a Geopath number and assume like-for-like — the methodologies differ.

Does Geopath measure digital and static the same way? The first seven steps are shared. Digital adds three more — rotation size, opportunity to see a specific spot in the loop, and spot-level noting — because a digital unit sells a share of time rather than the whole face.

Is a Geopath impression the same as a web impression? No. A web impression is a served ad, counted directly. A Geopath impression is a modelled estimate of a human being likely noticing a physical display. They are both called impressions and they are not the same unit. Any dashboard that adds them together is producing a number with no meaning.

Can a media owner inflate Geopath numbers? Not the methodology itself. But inventory selection, audience definition, share-of-loop assumptions and OTS-versus-adjusted framing all sit with the seller. That is where proposals get optimistic — not in the model.

Do I still need attribution if I have Geopath data? Yes. Geopath establishes that the audience was real. Attribution establishes that it responded. Advertisers increasingly expect both, as we covered in What Advertisers Now Expect From an OOH Campaign Report.

---

Sifra builds live measurement dashboards for fleet and OOH media owners — planned versus delivered, per unit, per day, with the modelled and counted layers kept honestly apart. See the Mobility vertical, or get a free mock dashboard built on your own campaign structure.