How out-of-home impressions are measured, and why moving screens are different

By Melih Koray, Marketing, Firefly · Published August 10, 2026

An out-of-home impression is a modelled estimate of how many people likely noticed an ad, not a recorded event. Geopath builds it from circulation past a display, reduced by a visibility adjustment. Most of those visibility inputs are properties of a fixed site, which is why moving inventory needs a different chain.

Out-of-home is the only major medium where the audience number is an estimate of human attention rather than a log of a server event. No one clicked. No device reported back. The number exists because a measurement body modelled it — and every model carries assumptions about the thing being measured.

Those assumptions are the subject of this page. They are also where mobile inventory quietly falls out of the standard framework, which almost nobody writes about.

What is an out-of-home impression?

Geopath, the audience measurement standard for out-of-home in the US, defines impressions as the total number of times people passing a display are likely to notice the ad, reported on a weekly basis.

Read that definition slowly, because two words in it carry the whole medium. Passing — the audience is the one in motion. Likely — this is a probability, not a count. An out-of-home impression is a modelled estimate of noticing, and treating it as equivalent to a digital ad impression is a category error that produces confident, wrong media comparisons.

How is the number actually built?

Three steps, each of which can be interrogated.

Step 1 — Circulation

The starting figure is Daily Effective Circulation. Geopath defines DEC as the total number of persons, in cars or other vehicles, passing and potentially exposed to an advertising display over a stated period — 12, 18 or 24 hours. It is assembled from vehicle counts, pedestrian counts and transit ridership for the location.

Step 2 — Visibility adjustment

Circulation counts people who went past. Most of them noticed nothing. The Visibility Adjustment Index is Geopath's correction: the percentage of a unit's total circulation likely to notice the ad. The glossary is explicit that VAI values come from visibility studies and reflect board size, roadside position, distance, road type and lighting — and that these are visibility factors, not audience measures in themselves.

Step 3 — From opportunity to likelihood

The industry used to report Opportunity to See: how many people had a chance to encounter the media. Geopath describes OTS as an outdated standard for out-of-home and reports Likelihood to See instead — the portion of the OTS audience likely to see the ad itself. That shift is the difference between counting who drove past a structure and estimating who registered what was on it.

Reach and frequency sit on top: reach is the share of a target population who notice the message at least once, frequency the average number of times they notice it.

Why does a moving screen break this chain?

Look again at what feeds the visibility adjustment: board size, roadside position, distance, road type, lighting. Four of those five are descriptions of a location. They are stable because the structure is bolted down.

Put the same ad surface on a vehicle and four of the five inputs stop being values and become distributions.

The circulation step inverts too. DEC counts people passing a display. For a vehicle, the display passes the people — so the equivalent question is not "how many go past this point" but "how many points does this unit go past, and who was at each of them."

What replaces those inputs for mobile inventory?

Structurally, four things do the work that location does for a fixed panel:

• Route — which areas the unit travels through, and in what sequence.
• Dwell — how long it stays within a given zone, including time spent stationary in traffic.
• Speed — how long any given viewer has the surface in view.
• Time of day — who is present along that route at that hour.

Notice what changes for the planner. On a fixed panel, geography is fixed at booking and cannot be adjusted. On mobile inventory, geography is an input you set — which means it can be planned deliberately, and also means an impression figure is meaningless unless the route it assumed is stated alongside it.

Frequency behaves differently for the same reason. A fixed panel accumulates frequency passively: the same commuters take the same road every morning, whether or not the campaign wanted that repetition. Mobile inventory can rotate routes to spread exposure across more people, or hold a route to concentrate it. Neither is better; they answer different briefs.

Is there a single standard for measuring moving inventory?

Not in the way there is for fixed panels. Geopath's glossary does recognise mobile formats — it defines mobile billboards as trucks carrying poster panels, either parked or driven through designated areas — but the visibility framework that produces a comparable impression number was built around stationary structures.

So the honest answer is that mobile impression figures vary by who produced them and on what assumptions. That is not a reason to discount the format. It is a reason to ask for the method before accepting the number, and to stop putting a fixed-panel figure and a mobile figure in adjacent columns of the same spreadsheet as though they were measuring the same thing.

What should you ask before accepting an impression number?

1. Who produced it? An independent measurement provider, or the company selling the space?
2. What were the inputs? Circulation source, visibility assumptions, time period.
3. Was the method designed for this format? Fixed-panel methodology applied to a moving unit is a mismatch, however precise the output looks.
4. For mobile: what route was assumed? Without it the figure has no geography, and geography is the variable.
5. Is it reach or impressions? They are different numbers and get quoted interchangeably by people who should know better.

Where to go next

For the formats these numbers describe and how each is bought, see the guide to out-of-home advertising. For who produces measurement and why the seller is rarely the right source for it, see how the out-of-home market is structured.