Sky AdSmart explained

A conventional television spot buys everyone watching a break. Addressable buys the households you selected, wherever they happen to be watching, by substituting the advert at the set-top box or in the stream.

That changes what you are buying from programme-and-time to household-and-attribute, and it changes the question you should ask about a proposal from 'how many people watched' to 'how well does this attribute set proxy my audience'.

It also means the effective cost per person reached depends on household size, which nobody controls and which is rarely mentioned.

The attribute catalogue is geodemographic: Experian Mosaic groups such as city prosperity and prestige positions, occupation attributes such as business owners and directors or senior decision makers in business, affluence indicators such as household income bands, and life-stage attributes such as young professionals or early technology adopters.

A brief naming a business audience compiles into the closest matching attributes intersected with the postcode districts for the cities you named, and mapped to broadcast regions. AdBuyMCP takes the top four matching attributes.

The honest reading of that is a household proxy. 'Senior decision makers in business' as an AdSmart attribute is a modelled household characteristic, not a verified job title, which is why the fidelity grade is household at 55 and why that score does not move however carefully the brief is written.

Entry campaigns have historically started around £3,000 per burst through resellers, on desk research dated 3 August 2026. That is the seller's minimum, and it is separate from any planner floor: AdBuyMCP enforces £1,000 on an addressable line, which stops a line too small to matter surviving into a multi-channel plan, and the sales house's minimum still applies at booking.

The market around it is moving. Comcast's Universal Ads launched UK self-serve in 2026 with low or no minimum, covering ITV, Sky and Channel 4 inventory, which is actively reshaping what an entry point looks like.

The buying process is a booking rather than an auction. A brief goes to the sales house, clearance happens, an order is confirmed, and delivery arrives afterwards as a statement.

Clearance is the binding constraint. Copy must be approved before it airs, superlative claims need documentary substantiation, and health claims are frequently blocked outright. Budget a fortnight and treat the clearance clock as the real dependency rather than media availability.

Production is a genuine cost too: a broadcast mezzanine at 1920×1080 and 25fps with at least 20Mbps, supers inside title-safe and audio to broadcast loudness standard. A social cut will not clear.

Concentration beats spread. Addressable television rewards a burst in a defined region; the same money spread nationally buys a frequency of roughly one, which by the usual planning convention is below the level at which television does much at all.

What it does: compiles your audience into the attribute vocabulary, intersects it with postcode districts and broadcast regions, scores the translation at 55 and says why, generates the booking brief with the right attributes, drives the clearance step, and holds the order state machine so every live step is acknowledged by a named person quoting the sales house's own reference.

What it does not do: make the booking, or report the delivery. In live mode the report pull throws rather than fabricating rows, and a pause requires sales-house acknowledgement rather than being assumed from a generated workflow.

Stated plainly: no addressable television delivery statement has ever been ingested through us. It automates the paperwork, which is a genuine saving in effort and is not the same as buying the media.