One model from content to proof.

Content, programming, airtime, screens and delivery evidence share one data model. Each step feeds the next instead of ending at a system boundary.

  1. 01Create

    Media, AI Studio, compositions, dynamic creative and live data.

  2. 02Program

    Playlists, channels, scenarios and mix policies.

  3. 03Sell

    Products, rates, bookings and guarantees.

  4. 04Operate

    Players, stores, zones and remote control.

  5. 05Prove

    Proof of play, delivery, reporting and settlement.

Transcript

The concept film shows the whole path in 73 seconds: create content, use AI Studio, program, sell, operate screens and prove delivery.

Create

Content that is made in the browser, adapts to data and passes review.

  • Dynamic creative

    One composition, many versions: rules on time, live data, place or scenario pick what each screen draws, and a store or zone sets its language. One ERP file can give each branch its own stock. Texts are machine-translated through DeepL with a brand glossary, the cost is shown before a run, and review can gate release. Every play records the version it showed.

    Details in the docs
  • Compositions

    Design scenes in the browser with widgets and live data; each is released as a versioned player bundle. A composition is the one content type that is created rather than uploaded. Scenes bind data fields and ship to players as offline bundles.

    Details in the docs
  • AI Studio

    Generate and edit images for your screens; every result is marked as AI and passes approval. Variant sets prepare versions for different screens, and provenance, approval and cost stay on record.

    Details in the docs
  • Media Studio

    Derive new assets from a ready image or video: wall parts, reframes, trims and stills. Nothing is encoded in the browser; one server-side job flow plans, encodes and reports progress for every operation.

    Details in the docs

Program

Playback that reacts to data, scenarios and demand.

  • Playlists & Channels

    Playlists loop ordered content; a channel binds playlists and paid demand to one screen area, with its own mix policy and scenario. A playlist can be narrowed per player category and time of day. Stores, zones and players carry category values such as placement or language, so one rule reaches many screens.

    Details in the docs
  • Scenarios

    One switch on a channel changes what plays, by hand, by API or by rules. Scenarios form one tree per workspace. Each channel stands on one node, so one venue area can run Football while another runs Wayfinding.

    Details in the docs
  • Mix policies

    One reusable policy splits a channel’s airtime between booked ads, programmatic sources and playlists, with a fallback for any gap. Stream targets add up to 100, a ceiling caps over-delivery, and each stream chooses what happens to airtime its demand cannot fill. One policy can back several channels.

    Details in the docs
  • Synchronised takeover

    The same motif at the same second on many screens, or one creative split across a video wall. Ordinary rotation does not start screens together; the docs catalogue what can be built, from which parts, and where each path stops.

    Details in the docs

Sell

Airtime that buyers can book, check and automate.

  • Bookings & targeting

    A product targets by channel, player, store or geo area, narrowed by categories; a booking adds flight windows and dayparts. The editor checks capacity as you edit, and advertiser clearance and operating hours remove what the screen cannot deliver.

    Details in the docs
  • Competitive separation

    Competing advertisers do not play back to back where another booking can serve; every exception is counted. The workspace keeps separation groups; an advertiser belongs to any number of them, and a booking may add its own.

    Details in the docs
  • SSP & Programmatic

    VAST sources fill the programmatic share of a channel’s mix, next to booked ads. Each ad opportunity runs one weighted draw, then an ordered waterfall. Demand Insights shows fill rate and latency per source.

    Details in the docs
  • Buyer agents

    A buyer’s own AI agent checks fit and price, drafts and requests bookings, and reads delivery and proof. It works on its own order book only, cut to the rights of its service-user API key.

    Details in the docs

Operate

One player engine and a fleet you can see.

  • Player platforms

    The same player runs as a native app or in a browser, including Samsung Tizen panels. Every target runs the same schedule, pacing and play-events; only the route onto the screen differs.

    Details in the docs
  • Stores & Zones

    Branches and zones organise the fleet, and a store’s opening hours and holidays apply to every screen in it. Hours are resolved in the store’s time zone and sent to the screens as exact instants, so a clock change cannot shift a holiday.

    Details in the docs
  • Monitor Players

    Fleet presence, player health and playback, with auditable remote commands. Simulate & Follow draws one screen’s programme at a chosen moment without a device.

    Details in the docs
  • Home dashboard

    Each person arranges Home from the cards they may see: which ones, their order and size, and their own tabs. Admins set a starting layout per role, a personal layout follows the account to any device, and cards never show more than the viewer’s permissions allow.

    Details in the docs

Prove

Delivery you can explain and verify.

  • Proof of play

    Finished plays are sealed hourly into a hash chain, anchored daily and, where a screen holds a key, signed by the device. A buyer can verify delivery independently instead of taking a number on faith.

    Details in the docs
  • Why this played

    Each play records why the spot won its slot, what the same decision denied and how rotation drew it. The play events are an append-only stream of what played, where and for how long.

    Details in the docs
  • Booking Health

    Can it be sold, can it play, is it keeping pace, did it keep its separation clause? Four questions at four moments, answered by the server in words.

    Details in the docs
  • Reporting

    Compare up to three bookings, players, channels or SSP sources, and read demand, freed airtime, carbon, proof anchors and versions played. A missing value always carries its reason instead of a dash, and most tabs export what they show as CSV.

    Details in the docs

Targeting

Seven levels. Combine any of them.

Geography, stores, zones, screen areas, categories, time and live data each carry a rule, and they stack: a screen plays only what passes every level that is on. Switch a level off to see what it removes.

SpaceRegion East › Retail. Decides who may see, book and approve this booking. Not where it plays.

Where it plays 2 of 24 screenssample estate

A data condition reads the screen’s latest snapshot; stale data falls back to safe content.

Details in the docs

Governance

Control without breaking the workflow.

Who may see, change and approve what is decided in six places.

  • Workspaces

    Every row belongs to one workspace. Reads are scoped to it and a write across workspaces is rejected.

    Details in the docs
  • Spaces

    One organisation tree per workspace. A role applies to a node and, optionally, below it; never upward.

    Details in the docs
  • Roles

    Permissions are named actions, bundled into roles. Advertiser memberships limit what a buyer sees.

    Details in the docs
  • Approval

    With content approval on, only approved content enters schedules. A new release waits while the approved version keeps playing.

    Details in the docs
  • Two-factor sign-in

    A workspace turns it on; people use an authenticator app or e-mail codes.

    Details in the docs
  • Operate as

    A platform operator may act as a user for at most 30 minutes, with a reason and no renewal.

    Details in the docs

Integration

Four front doors. One domain model.

Each door has its own credential. All of them call the same handlers, so workspace, permission and lifecycle rules exist once.

  • Browser · JWT

    Operator application

    REST and server-sent events for the browser application that carries the complete operator workflow.

    Details in the docs
  • Device · player token

    Player API

    SignalR control and read-only content delivery for paired players.

    Details in the docs
  • Customer · scoped API key

    Integration API

    Versioned REST for products, availability, bookings, media, delivery and settlement, with signed webhooks and a Scalar reference.

    Details in the docs
  • Agent · service-user API key

    MCP

    Typed tools on two paths: an operator path that also carries 23 resources, and a buyer path limited to the buyer’s own order book and the key’s rights.

    Details in the docs
Operator path · /mcp/operator
Show me players in Vienna that are offline or in maintenance, then explain what the affected screens last played.
Buyer path · /mcp/buyer · own order book only
Does the Spring launch product fit two weeks in March? Check fit and price, save the draft and request it.

Delivery quality

Know what will deliver, before the month ends.

Reach says how many screens could have shown a campaign. These readings say whether its promise is being kept. The server explains every shortfall in words, the same in the application, the customer portal and the API.

Pacing

On track, at risk or behind, per booking and per day, against the airtime owed so far. A booking that falls behind shows it mid-flight, while there is time to repair it.

Details in the docs

Shortfall, with its cause

A share gap can be repaired in the schedule. A volume gap is airtime that never existed, because the screens ran less than planned. The split says which to fix.

Details in the docs

Coverage

Where a saved booking is planned to land: eligible, cut out, crowded or closed, as a day strip or an hour grid. A hole in the plan is fixed before it becomes a shortfall.

Details in the docs

Settlement

Planned against measured seconds, per screen, as a post-flight export for finance. Price per thousand plays bills only plays that ran to completion.

Details in the docs

The detail is in the documentation.

Each block above links to the docs page that states the rule.

AdPrecision

A non-commercial research platform for digital signage and retail media.

© 2026 AdPrecision. No cookies, analytics or third-party requests.Independent research project. Not for sale.