Your China production partner delivers a product film for an overseas retail rollout. The retailer later reports ten thousand screen plays. Can your team divide the campaign budget by that number and call the result a cost per viewer?
The number is illustrative, not client data. Before calculating anything, establish what was counted: a file request, a rendered ad, or a person's opportunity to encounter it. Creative acceptance, media execution and audience results need different supporting records.
The industry already distinguishes playback from an opportunity to see
The December 2024 final IAB/IAB Europe in-store retail media document defines Ad Play as the number of displays or renders in a format. It distinguishes exposure-related measures and makes clear that an opportunity to see is not confirmation that someone watched.
IAB Europe's September 17, 2026 follow-up proposals emphasize footfall and visual contact opportunity rather than ad play alone, including transparent assumptions for modeled estimates. The announcement explicitly labels these as proposals, with comments due October 23. This article does not treat that proposal as an enacted replacement standard.
ONCE analysis: a retail video brief should specify which records establish each supplier's work. Successful playback cannot establish a viewer count; potential contact cannot by itself establish comprehension or incremental sales.
The ONCE four-receipt method
This is a purchasing and review method, not an IAB form or an automated measurement product. A small campaign can combine the records on one page while keeping their conclusions separate.
Receipt one: creative delivery
Identify the asset, product model, language, format, duration and approved version. Check the actual content and the recipient's agreed specifications. Name the environments used for playback testing. A successful test on a production computer does not establish compatibility with every retail device.
The production partner can accept responsibility for the agreed asset and adaptation work. Unless media operation is in scope, its export record cannot certify a complete store rollout.
Receipt two: actual playback
Ask the scheduling operator for screen or device identity, location, time period, asset version and the meaning of its recorded event. A planned playlist describes intended delivery. An execution record describes what the system reports happened.
Confirm whether the records distinguish opening hours, offline devices, wrong versions and incomplete plays. Systems differ in the detail they provide. Where the evidence is unavailable, record that limitation instead of treating missing data as successful execution.
Receipt three: audience opportunity
When a report supplies people or exposure estimates, ask which area and time period they cover, whether they are measured or modeled, and how assumptions and repeat contacts are handled. Whole-store entrance traffic should not silently become the viewer count for a screen beside one shelf.
An opportunity to encounter the ad is not a completed viewing or a brand-memory result. With limited audience evidence, report the known playback facts and leave the audience conclusion unresolved. Collecting more personal data is not a prerequisite for an honest report.
Receipt four: business outcomes
Report sales, scans, enquiries or counter interactions from the records actually available. Promotions, stock availability and merchandising changes may affect interpretation. Playback logs alone cannot assign a sales change to the creative.
If the question is what would have happened without the campaign, arrange an appropriate measurement design separately. Not every small project needs an extensive study. It does need to avoid claiming incremental impact when the supporting design is absent.
Locate the missing receipt before ordering a new edit
Consider a hypothetical two-store campaign. One screen runs in a staffed demonstration area; another device's totals include overnight loops. This is an example, not an allegation about a retail network or an observed ONCE project.
Equal play totals would not establish equal audiences. Lower sales at one location would not immediately justify a demand for a more dramatic opening. First establish actual delivery and the relevant hours, then examine audience evidence and the creative question.
- The file is wrong or fails in the agreed environment: resolve the specific production or adaptation fault.
- The asset is correct but scheduling or device status is unclear: investigate execution before reshooting.
- Playback is established but audience estimates are opaque: request methodology before comparing costs per viewer.
- Audience opportunity is supported but comprehension is unknown: consider a suitable creative review without pretending the sales effect is already isolated.
The method is not a way to move blame between suppliers. It makes the corrective action match the missing evidence, even when one integrated provider performs all the work.
Scope technical acceptance separately from effectiveness research
A production engagement may include concept, filming, screen variants, playback tests and defined revisions. A media engagement may cover scheduling, device operation and reporting. Audience or business-outcome research requires its own agreed scope.
Do not use the number of exports as a substitute for deployment coverage. Equally, a production partner cannot replace promised on-site adaptation with “we sent the file.” Specify what completion means for each responsibility.
For ongoing rollouts, retain screen requirements, approved assets, test records, playback identifiers and report definitions. A future refresh can then establish whether its records remain comparable. This is a proposed content service asset, not a claim that ONCE operates a sensor network or has demonstrated a particular efficiency gain.
Bring a report example into the production brief
Request a redacted report sample before commissioning the rollout. Identify planned values, execution records and estimates. A polished total-play figure alone is not enough to promise audience-based evaluation of a film.
When discussing retail product content with ONCE, bring the screen types, receiving specifications, known scheduling arrangement and an example report. Define the verifiable handover first, then decide which new creative assets the rollout needs.
Sources and scope
IAB/IAB Europe: In-Store Retail Media Definitions and Measurement Standards, final December 2024 version; and IAB Europe: In-Store Retail Media Measurement Proposals Open for Public Comment, September 17, 2026. Both checked October 30, 2026.
The sources support the metric distinctions and transparency principles. The four receipts and scenarios are original ONCE recommendations, not certification, a platform-specific measurement formula or a performance guarantee.
