DOOH CPM Explained: How the Number Is Built and How to Read It Honestly

Every digital out-of-home proposal opens with a CPM, and almost every one is read wrong. Buyers borrow the instinct from paid search and paid social — lowest cost per thousand wins — and apply it to a medium where the impressions underneath the number were never counted, only modelled. Two networks can quote the same CPM and deliver completely different value.

This post takes the DOOH CPM apart: what it is actually made of, where the modelling hides, and the small set of questions that separate a defensible rate from a decorated one. It is the money-layer sitting underneath everything we covered in Plays vs. Impressions: What You're Actually Buying in DOOH — that post deliberately declined to quote a benchmark CPM, and this one explains why that refusal is the honest answer.

What is CPM in DOOH?

CPM is the cost to buy one thousand impressions. In DOOH the mechanics are the same as anywhere else — spend divided by impressions, times one thousand — but the impressions are a modelled estimate of opportunity to see, not a counted event. As Captivate's breakdown of DOOH CPM puts it, in digital channels an impression is tied to a specific ad served to a specific user; in DOOH the impression is inferred from environmental and behavioural inputs. You are buying modelled exposure potential, not guaranteed views.

That single difference is why CPM behaves differently here. It remains a useful cost input. It is not, on its own, a measure of value — and treating it as one is where most DOOH evaluations go wrong.

How is a DOOH CPM actually calculated?

The formula is trivial; the denominator is the whole story. A DOOH CPM is campaign cost divided by estimated impressions, and those impressions are built from a stack of inputs: how often the ad plays, how many people pass or dwell in view, and a visibility adjustment for whether the screen could realistically be seen at all.

Walk it in order. A screen sells time in a loop, so your share of that loop sets how many times your creative plays. Each play is then multiplied by an audience estimate drawn from footfall or traffic data, dwell time, screen placement and orientation, and dayparting that weights exposure by hour of day, as Captivate and industry pricing guides such as The Media Ant's overview of DOOH pricing models describe. The result is an impressions figure — and because that figure is modelled, the same play count can produce very different CPMs depending on which multiplier the seller applied.

This is the same impression multiplier we unpacked in Plays vs. Impressions, viewed from the cost side. If you do not know the basis of the impressions, you do not know the basis of the CPM.

Why can a cheap CPM be the expensive one?

Because a low CPM is often low for a reason: it is counting impressions generously. A screen that models every passer-by as a viewer, applies a flat multiplier across all 24 hours, and reports raw opportunity-to-see will show a large impressions number and therefore a small CPM — while delivering far less real attention than a higher-priced, visibility-adjusted alternative.

Captivate frames this as two traps that mislead in opposite directions. Cheap inventory can hide screens that are technically in view but easy to ignore, short dwell times, and minimal verification that the ad ran at all. A higher CPM can reflect the opposite — controlled environments, longer exposure, verified delivery — and be the more efficient buy per unit of actual attention. The price tag alone tells you neither.

The practical consequence for a media owner: a low headline CPM is not a selling point if you cannot show what is underneath it. A rate you can defend line by line beats a cheaper rate you cannot.

Visibility-adjusted impressions or raw opportunity-to-see?

This is the single question that moves a CPM the most. Opportunity-to-see (OTS) counts everyone who could have seen the screen. A visibility-adjusted impression counts those estimated to have actually been able to notice it, after correcting for angle, distance, illumination and dwell. The gap between the two can be a large multiple — which means the same price is cheap on one basis and expensive on the other.

We walked the full ten-step chain that narrows raw circulation down to a noted impression in What Is Geopath, and How Is an OOH Rating Actually Built?. The takeaway for CPM is direct: a CPM computed on visibility-adjusted impressions and a CPM computed on raw OTS are not the same product, and comparing them side by side without saying which is which produces a meaningless ranking.

What is a good CPM for DOOH?

There is no single honest benchmark, and any number quoted as one should be treated with suspicion. DOOH CPM varies too widely by market, format, venue, audience definition and — critically — impression basis to reduce to a figure. A premium transit screen and a low-traffic roadside panel are not the same buy, and a CPM on A18+ totals is not comparable to a CPM on a narrow demographic target.

This is not evasion; it is the same data-first stance the whole cluster is built on. The useful question is never "is this CPM low?" It is "what is this CPM computed on, and can the seller show me the basis?" A defensible methodology at a higher CPM is worth more than an attractive number with nothing underneath it.

How should CPM sit on a campaign dashboard?

Show CPM as an output of two labelled layers, never as a standalone headline. Put the modelled impressions on one line, the verified delivery — plays, screen uptime, share of loop — on another, and let CPM be derived from them where the viewer can see the inputs. The number a CFO trusts is the one whose denominator is visible.

The metric that actually earns renewals is the effective CPM tracked over the flight: cost against visibility-adjusted impressions, moving week to week, next to the delivery proof it rests on. When effective CPM drifts, the dashboard should show whether it moved because delivery changed or because a modelling assumption changed — a distinction that matters as much as the number itself, and one we made the core of What Advertisers Now Expect From an OOH Campaign Report.

FAQ

Is DOOH CPM the same as digital CPM? No. A digital CPM rests on counted, served impressions. A DOOH CPM rests on modelled opportunity-to-see. Both are called CPM; the denominators are different kinds of number and should never be pooled into one blended figure.

Why do two screens in the same area have such different CPMs? Because DOOH has no standardised rate card, and because impression bases differ. Placement, dwell, share of loop and the visibility adjustment all vary screen to screen, so identical locations can price very differently and both be legitimate.

Does programmatic DOOH fix CPM transparency? It helps on buying flexibility and can tie price to real-time audience conditions, but it does not by itself standardise how impressions are modelled. The basis question — visibility-adjusted or raw OTS — still applies to a programmatic buy.

What should a media owner disclose about their CPM? The impression basis, the audience definition, the share of loop, and whether delivery is verified. A CPM offered with those four things stated is a rate a buyer can defend internally; a CPM offered without them is a number waiting to lose an argument.

Should I just pick the lowest CPM? Only if you have confirmed the impressions underneath are on the same basis as the alternatives. Otherwise the lowest CPM is often the least honest denominator, not the best value.

---

At Sifra, this is the Mobility work: campaign dashboards that show CPM as the output of labelled delivery and modelled-exposure layers, so the rate is one an advertiser can actually trust. Data, made visible.

Curious what yours would look like? We'll build you a free mock campaign dashboard shaped around your own inventory — no commitment.