Choosing between cinema and CTV

Connected TV selects households, through consent-chained marketplace segments layered over geography with a frequency cap. Where at least one segment matches your audience strongly the plan is graded person-level at 85; where matches are weak it drops to household at 65; where nothing matches it falls back to geo-city targeting at 60 rather than pretending the segment was found.

Cinema selects a venue and a film. Audience packs and genre packs combine with a venue list, and the resulting plan is a projection from catchment demographics onto a screen. It scores 40 out of 100, venue-level, and that number is fixed — no amount of brief quality moves it.

The gap between 85 and 40 is not a quality judgement about the channels. It is the honest distance between what you asked for and what each one can be sold, and it should change the claim you make afterwards rather than necessarily the decision.

A cinema audience is seated, in darkness, with no second screen and no skip button, watching a screen that fills their field of view with sound they cannot turn down. No other advertising environment in a UK media plan comes close on attention per exposure.

That makes cinema an impression instrument rather than a frequency one. It is bought for the quality of a small number of exposures, not for the accumulation of many, which is why it works for a launch and rarely for ongoing performance.

It is also a small and volatile market — around £220m in the UK in 2025 — which means it cannot absorb a large budget and its availability swings with the release schedule.

Connected TV can be live in days. Creative goes through a pre-check, trafficking metadata is attached, and delivery starts.

Cinema takes around eighteen days from brief to on-screen once clearance sits in the path, and it is booked as paperwork rather than bought at auction: a brief goes to the sales house, clearance happens, an order is confirmed. It also needs a different master — a 2K flat package at 24fps with 5.1 audio, sixty seconds maximum — rather than a broadcast mezzanine or a social cut.

In practice this is the constraint that resolves most of these decisions. A campaign that has to be live in ten days cannot include cinema, whatever the merits.

Connected TV reports impressions and can be measured on observed response through your own analytics on the campaign's immutable key. Its geography usually supports a powered geo-lift test if the spend and conversion volume allow.

Cinema reports admissions against a venue list, delivered as a statement rather than as an API response. There is no person-level exposure record and no UK seller can provide one. Geo-lift is the credible causal design, and it needs enough separated catchments to work.

Neither can be measured properly on last-click, and cinema in particular will look like it did nothing under an attribution model that credits the final touchpoint.

Cinema targeting involves no personal data at all: audience packs, genre packs and venue catchments are aggregate. The compiler records it as outside UK GDPR scope for delivery rather than leaving it to be inferred.

For a regulated category, or an organisation whose data protection posture makes behavioural targeting difficult, that is occasionally the deciding factor rather than a footnote.

Cinema has its own clearance body and its own age rating, though, and copy with strong language restricts pairing to 15 and 18 certificate films — which can invalidate the venue package the plan was built on. The regulatory advantage on data does not extend to content.