4 Oct 2026

What in-room TV ad inventory is actually worth: fill rates, CPMs and the seasons that move them

Operators routinely overestimate one of the two numbers that decide in-room ad revenue. A straight look at how ad buyers price in-room TV inventory, what fill rates are realistic, and how seasons move both.

Every in-room screen model eventually reduces to three numbers: how many hours the screen is watched, what share of those hours you can sell (fill rate), and what an advertiser pays per thousand of those views (effective CPM). Our earlier piece on what a Smart TV can actually earn per room set the ceiling. This piece is about the two numbers underneath the ceiling, because when operators are disappointed by in-room ad revenue, it is almost never the ceiling. It is the fill rate.

Why fill rate is the number nobody quotes

Fill rate is the share of available ad slots that actually sell. A screen with 200 sellable slots a month that sells 60 of them has a 30 percent fill rate, and no amount of CPM optimism fixes that arithmetic. New programs start low. An inventory pool of 40 rooms in one building is a niche buy; national programmatic buyers want aggregated reach in the thousands of screens before they treat your inventory as a line item in a plan. Realistic year-one fill rates for a single-property program live in the 20 to 40 percent range, and the honest ones get there through local and regional advertisers, property-adjacent categories like airport transfers, furniture rental, tours and delivery services, not through national campaigns.

This is why aggregation matters more than any single property's occupancy. Ten operators pooling 400 rooms under one inventory code become sellable in a way that 40 rooms never are. If you are a single operator, the practical question for any vendor is not "what CPM do you pay" but "how many screens are in the pool my rooms join."

What in-room TV CPMs look like in practice

Standard digital-out-of-home video inventory trades broadly between AED 15 and AED 60 per thousand impressions depending on targeting, dwell quality and proof-of-play. In-room short-stay inventory sits in the interesting middle: the audience is captive, financially qualified and watching from two to four metres, which is better attention than an elevator screen, but the impressions are hard to verify without per-screen heartbeat data. Buyers who pay premium rates do so for exactly that verification. A program that can produce per-screen play logs and honest uptime numbers, the disciplines we covered in the uptime piece, defends its rates. A program that reports scheduled impressions gets repriced as unverified inventory, which lands it in the low teens fast.

Two categories consistently pay above the average for short-stay inventory: travel services that convert at the moment of the stay (transfers, late checkout, excursions) and home services with local intent. Financial products pay well but demand compliance review and audience data most short-stay programs cannot supply yet. Retail brand advertising pays the least and fills the most, which is why it anchors year-one fill.

The seasonality nobody models

Ad demand is seasonal in ways that roughly track your occupancy, which is a cruel coincidence. Peak season is when you have the most guests watching, and also when buyers face their own budget freezes and post-budget quarters. The months when a Dubai or Mediterranean short-stay property sits at 35 percent occupancy are frequently the months when local businesses are spending hard on capturing the visitors who did come. Low season is when the in-room model gets stress-tested, and it usually fails there for a predictable reason: the operator priced the program on peak-season revenue and has no answer for the trough.

The answer is not better ad sales in the trough. It is weighting the model toward the side of the screen that does not depend on third-party buyers: the in-room upsell menu. Room upgrades, early check-in, parking and breakfast attach to the guests who are actually in the building, and a guest in a 35-percent-full building in August is exactly as monetisable as one in December. The mix shifts seasonally, and the operator who treats the screen as an upsell channel with an advertising layer on top survives February; the one who treats it as an ad screen with upsells as a garnish does not. How to build that menu from your own request log is in the offer menu piece.

How to read a revenue report you can trust

Ask for four figures per property, per month: sellable hours delivered, fill rate, effective CPM, and revenue net of any split. Effective CPM, meaning total revenue divided by verified plays times a thousand, is the only CPM worth discussing; quoted rate cards are fiction until they survive a month of delivery data. If a vendor reports only scheduled impressions and a rate card, you are looking at a forecast pretending to be an invoice. The split itself, who holds the money between you, the operator and any management company, is a contract question we set out in the revenue split piece, and it is worth settling before the first dirham arrives, because retroactive accounting on ad revenue is where these programs quietly leak money.

The realistic shape of a first year for a mid-size property is unglamorous: fill rate in the 20s rising toward 40 as the pool aggregates, effective CPM in the low-to-mid tens of dirhams, upsells carrying more than the advertising in the weak months. That is not a failure state. That is a working program compounding. The failure state is a single-property inventory pool nobody can sell, reported in scheduled impressions, priced off a peak-season spreadsheet.

Running screens across multiple properties? AdVision Connect reports fill rate and effective CPM per property, so you know what your inventory actually sold for, not what it was scheduled for. Talk to us about a pilot.