DOOH advertising: how the impressions are counted

One measured play, multiplied by a modelled estimate.

If you buy Meta and Google for a living, a DOOH deck will feel familiar. Impressions. Audience segments. Programmatic. Dayparting. Almost none of those words mean what you think they mean.

Digital Hangover does not resell screens. We sit on the client's side of the table when a screen vendor arrives, which is the job this page is written for. For how DOOH sits beside everything else you buy, see our overview of marketing channels.

What is DOOH advertising, and how is it different from a hoarding?

DOOH stands for digital out-of-home: advertising on an internet-connected digital screen in a public or shared space — a mall LED wall, a metro platform screen, an airport display, an office lift panel, a cinema screen before the trailers.

Classic out-of-home is a printed hoarding. You book the site for a fortnight, the vinyl goes up, it stays up. DOOH changes four things at once.

Switching a message by daypart is how DOOH supports moment marketing and festive campaigns in a way vinyl never could.

Nor is out-of-home shrinking in India. The Pitch Madison Advertising Report 2026 put Indian OOH spend at ₹4,835 crore in 2025, up about 4% year on year, per exchange4media. Screens take a growing share of that, but no dated source we could find splits out the digital portion, so we will not put a number on it.

How is DOOH bought in India?

Two routes, and the difference matters.

Direct with the media owner. You negotiate with whoever operates the screens — a metro concessionaire, a mall network, a cinema chain, an airport media company. You buy a fixed number of spots per loop per day across named sites, and you get a site list, a schedule and, if you ask, proof of play.

Programmatic DOOH via a DSP. The same screens go into an exchange and you bid for individual plays from a demand-side platform. The plumbing is your display plumbing: the IAB Tech Lab and the OAAA extended the OpenRTB protocol to carry DOOH bid requests, as AdExchanger covered in December 2022. We are not re-teaching that machinery; our guide to programmatic advertising in India covers DSPs, exchanges and auctions, and it applies directly.

What changes between the two:

• Commitment. Direct means a fortnight and a site list. Programmatic lets you buy a week, or a weather trigger.
• Price. Programmatic adds platform and exchange fees on top of media. You pay for flexibility, not a discount.
• Transparency. Direct gives you named sites. Programmatic often gives a venue type and a city.
• Targeting. Programmatic layers audience and location rules — with the caveat below.

Google's documentation for DOOH in Display & Video 360 is unusually blunt here: DOOH geo-targeting leans on IP addresses, and the IP a publisher passes may describe its own server rather than the screen, so some plays can serve outside your target area. If precise location is the whole point, prefer a named site list. Our note on geo-targeting covers the principle.

How is a DOOH impression actually calculated?

This is the part the deck skips. A DOOH impression is not a count of ad deliveries. It is plays multiplied by a modelled estimate of how many people were likely to see each.

DOOH impressions = number of plays x impression multiplier

impression multiplier = modelled audience for that screen
                        in that 15-minute slot, per play

The clearest public definition of that multiplier comes from the UK, where the trade body Outsmart publishes an Impression Multiplier Standard (February 2026). The method: take the measured audience for a screen over a period, divide by the spots played, and you have average audience per play. Their worked example is 150 audience impacts across 15 plays, a multiplier of 10. Multipliers are produced per screen in 15-minute blocks — 8,064 of them for one screen across the week and month.

Where does the audience figure come from? Modelling. In the US, Geopath's published methodology builds its numbers from road-network data, points of interest, trip modelling, demographics and a mobile-device location panel. Nobody counts heads at the screen.

The IAB's July 2025 explainer names two tiers plainly: Opportunity to See, meaning a person was somewhere the screen was visible, and Likelihood to See, a smaller number adjusted for whether they plausibly looked. Most impressions you are quoted are closer to the first.

Working through the maths — the calculation we run whenever a vendor quotes an impression number, on a deliberately hypothetical screen:

• Your creative is 10 seconds in a 60-second loop. Six plays an hour, and a 16.7% share of voice.
• The screen runs 14 hours a day: 84 plays per screen per day.
• The vendor's multiplier for that location and daypart is 10, so 840 impressions a day from one screen.
• Across 100 screens, 84,000 impressions a day — an impressive number in which the only hard figure is 84.

The plays are logged. The 10 is an estimate. If the multiplier is 30% optimistic, so is your CPM, and nothing in the report will say so.

Why a DOOH impression is not a digital impression

A Meta or Google impression is a logged event: a device requested an ad and it rendered. A DOOH impression is a forecast of human attention attached to a confirmed screen event. They share a word, nothing else.

Display & Video 360 separates the two in reporting, and the distinction is worth stealing. Served impressions count how often your ad appeared on a screen. Publisher impressions apply the publisher's multiplier to estimate viewers — and Google states outright that the multiplier is the publisher's estimate of foot traffic, not a verified count, and that device-level impression measurement does not apply to DOOH.

Three consequences:

• No click. No clickthrough rate, no session, no bounce. A QR code gives you a scan rate from a tiny slice of viewers.
• No cookie or device ID at exposure. The screen does not know who looked — which is also why DOOH is comparatively clean on privacy.
• No shared currency in India yet. The IOAA has been backing an audience measurement platform, and its chairman told exchange4media in June 2026 that the sector is mid-transition towards TV-grade rigour. Until then, two vendors can quote the same screen differently and both be "right".

If you want a screen medium that does measure like digital, it is connected TV — see CTV advertising in India and Hotstar ads. Do not expect DOOH to behave that way.

What can you actually measure on a DOOH campaign?

Four real methods. Every one is an inference — say so when you present the results.

The strongest is a holdout design, because it answers the only question that matters: what would have happened anyway. Same logic as incrementality testing, and at large spends marketing mix modelling is where DOOH gets a fair hearing. Last-click reporting shows DOOH contributing nothing — a flaw in the method, not the channel, as our explainer on marketing attribution sets out.

Where DOOH earns its place in an Indian media plan

DOOH works where attention is captive, the environment is premium, and you need to be believed rather than clicked. It is a top-of-funnel and brand awareness instrument.

• Metro networks. Long platform dwell, daily repetition, a defined commuter audience.
• Airports. Expensive, and the clearest credibility signal available to a young brand.
• Malls and retail corridors. Proximity to purchase — the natural pairing with retail marketing.
• Cinemas. A seated audience and the longest attention in the medium.
• Office parks and lift screens. Tight B2B targeting by building, not by interest graph.
• Transit corridors and roadside LEDs. Mass city reach, weakest measurement of the set.

Cost is driven by six levers: the location's audience tier, your share of the loop, creative duration, the dayparts bought, the screen's format and size, and direct versus the programmatic premium. We will not quote rates — two malls in one city can differ several times over, and any rate card online is already stale.

It is a poor fit if you need measurable leads this quarter, or if your audience is defined by behaviour rather than geography. There, a hyperlocal marketing approach reaches the same catchment for far less. Brief the creative for three seconds and six words: unlike feed media buying, you have a glance, not a scroll.

The questions we put to a DOOH vendor before a client signs

First-hand, from our own client work: we do not book outdoor media, but we review the proposals. These are the eight questions we send back, and the answer that makes us advise walking away.

Two decide the deal. Proof of play is the only hard evidence that your money bought anything, so it belongs in the contract, not the pitch. And playing screens versus listed screens is where money disappears: a network sells 300 screens, 240 are live this month, and your forecast was built on 300. Get the live count in writing, dated.

Buying screens with your eyes open

DOOH does not need defending. Captive attention in a premium environment is scarce, and a metro screen reaches people with ad-blockers and no patience for feed ads. What it needs is honest arithmetic. Treat the impression number as a forecast, the play count as the fact, and a holdout as the only proof. Judge it on whether branded search, direct traffic and store visits moved in the exposed cities — not on a CPM that was never comparable to your Meta CPM. That is the frame we use across every channel in performance marketing, and screens are no exception.