In-Store Retail Media Measurement.
An independent decode of the documents currently served only by primary PDFs and vendor blogs — the full impression ladder with its modeled-exposure caveats, zone-by-zone guidance, the sales rules, and the maturity framework.
In-store retail media measurement is defined by two IAB documents: the joint IAB and IAB Europe In-Store Retail Media Definitions and Measurement Standards, final December 3, 2024, and the Viable Framework for Maturing In-Store Media Measurement, December 9, 2025. Together they set store zones, an impression ladder, and sales-measurement rules.
Last validated July 13, 2026 · Maintained by Evgeny Popov
The most load-bearing sentence in the standards is a modest one: today, networks should report Ad Play and Gross Impressions with formulas disclosed. Everything above that floor is aspiration — and the top rung carries the document's entire privacy treatment in a single line.
Fast read
- What it is
- A deep-dive decode of the two documents that govern in-store retail media measurement: the joint IAB and IAB Europe In-Store Retail Media Definitions and Measurement Standards (final December 3, 2024) and the IAB’s A Viable Framework for Maturing In-Store Media Measurement (December 9, 2025).
- What it covers
- The five store zones and their placements, the four-rung impression ladder with the standards’ own evidence annotations, exposure zones and traffic metrics, the worked OTS calculation, the sales-lift and incrementality rules, and the maturity framework’s layered adoption path.
- What it is not
- Not the overview. The Retail / Commerce Media Measurement hub places these documents inside the five-document retail corpus; this page goes inside them. The 2024 standards also scope themselves to digital screens, audio, and connected shopping only — print, static, and experiential formats are deferred to future iterations.
- The buried headline
- The most load-bearing sentence is modest: today RMNs should report Ad Play and Gross Impressions, with formulas disclosed. Everything above that floor is aspiration — and the top rung, LTS, carries the standards’ entire privacy treatment in one line: "NB: This might have GDPR implications."
- Best for
- RMN measurement leads standing up in-store reporting, brands and agencies auditing an in-store proposal, and vendors deciding which rung of the ladder their technology can honestly evidence.
- Best next read
- Retail / Commerce Media Measurement for the full five-document corpus these standards sit inside; DOOH & Place-Based Media for the shared physics of screens in places.
Positioning note: the Retail / Commerce Media Measurement hub covers in-store at overview level, inside the five-document retail corpus. This page is the deep dive — if you want the corpus map, start there; if you want the ladder, the zones, and the rules, stay here.
Two documents, one lineage.
The December 2024 standards are a joint IAB Europe and IAB (US) publication — 35 pages, final December 3, 2024, after a September–November 2024 public comment period, originating in a July 2024 IAB Europe workshop with fourteen RMNs and IAB US's May 2024 DOOH/In-Store Playbook. They build on the IAB Europe Retail Media Measurement Standards (April 2024), the IAB/MRC Retail Media Measurement Guidelines (January 2024), and the MRC Digital OOH Measurement Guidelines — which the document says were used as the basis, but needed refinement because stores are tied to a point of sale. Participants include Walmart Connect, Kroger Precision Marketing, Albertsons Media Collective, Ahold Delhaize, Tesco's Dunnhumby, REWE, Schwarz Media, SMG, Instacart, Roundel, CVS, Walgreens, and Home Depot, with MRC involvement. Scope, for this version: digital screens, audio, and connected shopping formats — print, static, and experiential are appendix examples with standards deferred to future iterations.
The citation trap
A September 2024 public-comment draft of the standards still circulates at an iab.com URL, and several vendor blogs cite it as if it were the standard. The governing text is the FINAL of December 3, 2024, hosted by IAB Europe — linked in the sources below. If a vendor's definitions do not match this page, check which PDF they read.
The second document is gated
A Viable Framework for Maturing In-Store Media Measurement (IAB, December 9, 2025) sits behind a free IAB account with no public PDF. This page's framework section is built from the IAB's own landing-page summary, triangulated against independent secondary writeups, and labeled accordingly — exact phase naming inside the full text is not publicly documented. Verify against the gated document before quoting its vocabulary in anything contractual.
The five store zones — and what each is good for.
The standards divide the store into five zones, each with named placements. The zone map is modeled — the document says so itself — on a larger big-box store with generalisations, and retailers can add nuances. The placements column below is faithful to the standard; the practical read is this page's decode.
| Zone | Placements the standard lists | The practical read (ours, not the standard's) |
|---|---|---|
| Zone 1 Exterior to Store | Charging stations, ATMs, curbside, drive-through, exterior building walls, fuel stations, shopping centres, lighting posts and fixtures, parking lots, window displays visible from outside. | Functionally DOOH bolted to a retailer: exposure happens before any shopping intent forms. Ask which counter feeds the audience number — parking or curbside traffic is not store footfall, and the exposure zone is hardest to define here. |
| Zone 2 Entrance and Out of Category | Security barriers and entrance-gate covers, foyer displays, service kiosks (self-service terminals, vending, tobacco kiosks), building walls, stand-alone displays, post-checkout and store exit, power aisles. | Highest traffic, lowest category intent. Door counters make entrance counts the cleanest denominator in the store — expect the most defensible Gross Impressions in this zone, and the weakest claim that exposure met a shopper mid-decision. |
| Zone 3 Check out | Belt and checkout surfaces, in-queue screens, shopper-facing transaction screens at POS, self-service checkout. | Captive dwell: queue time makes dwell-time claims most credible here. But the basket is already built — checkout exposure supports next-trip and brand effects, not same-basket lift, and a lift methodology should say which one it is claiming. |
| Zone 4 In Aisles | Prime promotional areas such as power aisles and gondola ends — one sentence in the standard, the thinnest zone description in the document. | The zone closest to the point of decision gets the least definitional support. Exposure-to-purchase claims are strongest here and instrumentation guidance is thinnest — the aisle-penetration and compliance multipliers in the OTS formula do their heaviest lifting in this zone. |
| Zone 5 Other & Connected Store | Third-party and auxiliary services (the standard’s explicit catch-all: bank branches, coffee shops, optometrists, or any other service between checkout and exit), BOPIS pickup, customer service, smart carts. | A deliberate catch-all plus the connected frontier. Smart-cart and BOPIS placements are device-logged — Ad Play evidence is strong — but the audience semantics differ: a smart-cart render is closer to onsite digital than to a screen in space. |
A quirk worth knowing
Power aisles appear in both Zone 2 (Entrance and Out of Category) and Zone 4 (In Aisles) — the standard itself is ambiguous about where they live. If a network's zone reporting matters to your buy, ask which zone its power-aisle inventory is filed under, because the answer changes which traffic denominator applies.
The impression ladder, rung by rung.
Four terms, four progressively stronger claims about the same playout, each annotated in the standard with its own evidence level. None are interchangeable; none are summable. The definitions below are faithful to the standard; the caveat column is where the modeled exposure hides.
| Rung | Definition (per the standard) | Evidence level (the standard's own) | Where the modeled exposure hides |
|---|---|---|---|
| R1 — Ad Play | The number of times an ad is displayed or rendered on a particular format. A device log event. | Ad play (digital) | No audience claim at all. It is proof the screen played the file — the in-store analogue of a served, not rendered-to-a-person, impression. One of the two metrics the standards say to report today. |
| R2 — Gross Impression | The number of individuals present in the defined Display Exposure Zone while the display is functional, over a period of time. Formula stated in the standard: ad impression = audience × ad play. | Presence in the Display Exposure Zone | The standard says it directly: this metric does not necessarily mean the ad was seen. The audience term is usually modeled — from door counters, sensors, panels, or transaction data — so the formula inputs, not the formula, carry the uncertainty. Disclosure of both is required. |
| R3 — Opportunity To See (viewable impression) | A single opportunity to view an advert: the number of people who could pass by an in-store activation — or, for in-store radio, who could hear it. The standards stress that opportunity is the key word: entering the store gave the individual a chance to see the ad, not evidence they saw it. | Viewability condition exists | The standards position OTS as the best proxy available for a viewable ad impression in-store, for digital activations — computable via a zone benchmark calculation or a measurement technology vendor. The benchmark route multiplies six or more modeled assumptions; the worked example below shows exactly which ones. |
| R4 — LTS Impressions (Likelihood-To-See) | A further refinement of viewable impressions: an adjustment for the likelihood individuals actually noticed or saw the content, determined using sensor or analytic technology. | Evidence ad has been seen | The frontier rung — and the one place the standards mention privacy, in a single appended line: "NB: This might have GDPR implications." Sensor-based attention measurement of shoppers is a personal-data processing question, expanded below. |
The what-to-report-today ruling
The standards are unusually direct: "Today RMNs should report on ad play and gross impressions." The formulas used must be disclosed, and for digital activations Opportunity To See is named the best proxy available for a viewable ad impression. Read as a hierarchy: floor = Ad Play + Gross Impressions with disclosed formulas; near-term aspiration = OTS; frontier = LTS. A network reporting only "impressions" with no rung named is below the floor.
Exposure zones, traffic, and the worked OTS example.
Every rung above Ad Play depends on an audience denominator, and the standards define where those denominators come from — and require the method to be disclosed.
- Display Exposure Zone: the area in which exposure to the video can be determined — with no assurance the ad can be seen. The standards also call this the Visual Exposure Zone.
- Audio Exposure Zone: the area in which the audio can be heard, with no assurance it is audible — and audibility defaults to an exposure zone of the entire venue unless the retailer notes otherwise. An in-store audio impression is, by default, a whole-store claim.
- In-Store Traffic: the number of measurable shoppers in-store, located at various zones — total store, entrance, near screens, shelves, aisles, checkout. Footfall may come from transaction data, panels, or technology such as sensors, door counters, and beacons; the collection method must be disclosed.
- Dwell Time: time spent within view of the advertising display, indicating the opportunity to engage. Retailers should provide aggregate and average dwell plus a distance matrix of viewability by zone — ask for the matrix, not just the average.
- The counting hierarchy: per the IAB/MRC basis documents, venue traffic ≥ screen traffic ≥ screen audience estimates ≥ average ad-unit audience estimates. Each successive count can never exceed the one above it — a quick arithmetic check that catches inflated audience claims before any audit does.
- Loop vocabulary: Loop, Ad Loop Duration, Ad Segment (the commercial pod), Ad Unit, and Ad Unit Length — the scheduling grammar that determines share of voice in the OTS calculation below.
The standard's own worked example: 12,800,000 OTS.
The standards include a zone-benchmark OTS calculation, created with input from SMG and Scala/PRN. It is the single most useful table in the document, because it shows exactly which modeled assumptions multiply into the headline number.
| Input | Example value |
|---|---|
| Average weekly store footfall | 50,000 |
| Number of stores | 100 |
| Campaign length (weeks) | 4 |
| Aisle penetration (share of visitors reaching the placement) | 80% |
| Share of voice in the ad loop | 100% |
| Touchpoint compliance (screens actually live and playing) | 80% |
| OTS per visit | 1 |
| Placement discount factor | 1 |
| Opportunity To See (product of all inputs) | 12,800,000 |
Note what the two 80% factors do: aisle penetration and touchpoint compliance together cut the raw footfall product by 36% before the number ships. Those are the honest levers — and the ones to ask for first, sourced, when a network quotes a benchmark OTS. A vendor-measured OTS replaces this arithmetic with sensor data; the disclosure requirement applies either way.
Sales measurement: windows, lift, and the method hierarchy.
The standards recommend sales be reported for 30 days pre-exposure, the duration of the campaign, and a 30-day post-exposure window as standard — with retailers able to provide flexible windows when brands ask. Sales Lift covers the advertised product; Brand (Sales) Lift covers the total brand within the same category — the equivalent of Halo in IAB Europe's online standards — and both must exclude non-activated stores. Sales Extrapolation, estimating impact on non-identified shoppers from identified loyalty shoppers, is permitted with the extrapolation basis disclosed.
- 01
Test vs. Control
The preferred method: randomized activated vs. non-activated stores, with group sizes and statistical significance reported. The standards’ store-matching best practices: matched pre-period sales, control for other in-store activity, identical price and promotion, same store format and size. Every one of those is checkable — ask for the matching table.
- 02
Multivariate testing
The fallback when external factors cannot be controlled well enough for a clean test-control split — modeling multiple variables at once to isolate the media effect. More assumptions than a randomized test, fewer than a pure model; the assumption list is the disclosure to demand.
- 03
One-to-One
Exposure-to-transaction linkage at the individual level. The standards call it the most accurate — and note it requires heavy technology investment and has scale and privacy limitations. In-store, individual-level exposure means sensors or identity at the shelf edge: the GDPR question returns at full volume.
The demotion that matters
Sales Variance — simple before/during/after deltas with no control — is explicitly demoted: it "should be used only when Incremental Sales studies are not possible." Most in-store case studies in circulation are Sales Variance studies. Under the standards, that is the last-resort method, not the headline one.
| Purchase cycle | Standard's examples | New-to-Brand / New-to-Category lookback |
|---|---|---|
| Regular | Shampoo, milk | 0–6 weeks |
| Semi-regular | Jeans, mascara | 7–26 weeks |
| Infrequent | TVs, sofas | 27+ weeks |
The purchase-cycle windows define what "new" means in a New-to-Brand claim — a 6-week lookback and a 26-week lookback produce different rates for the same campaign. The IAB says it will aim to standardise product categories in the next version; until then, the window is a disclosure to demand, not an assumption to make.
The maturity framework: what stores can adopt today.
A Viable Framework for Maturing In-Store Media Measurement (IAB, December 9, 2025) exists because — per the IAB's own summary — in-store adoption has been slowed by operational complexity, inconsistent standards, and a lack of comparability across networks. It offers a standard measurement baseline that retailers and vendors can adopt today and an actionable path grounded in current capabilities. As reported by independent secondary writeups, the structure is three layers, not numbered phases. It is a framework, not an IAB/MRC standard — vocabulary should come from the December 2024 standards text.
- Layer 1
Baseline — verified impressions
A standard measurement baseline that retailers and vendors can adopt today, per the IAB’s published summary. Secondary writeups describe the baseline metric as verified impressions — note that this is not one of the four defined terms in the December 2024 standards, so pin the definition to the framework’s full text before adopting the word.
- Layer 2
Insights — planning language
Unique reach, frequency, and dwell time — the familiar planning vocabulary of every other channel, brought in-store. As reported by independent secondary writeups; the IAB’s own summary frames the framework as an actionable path grounded in current capabilities.
- Layer 3
Outcomes — closing the loop
Linking in-store exposure to sales, with a stated preference — as reported — for deterministic linkage over probabilistic modeling, and disclosure required where probabilistic methods are used. This is the layer where the in-store ladder meets the sales-measurement rules of the 2024 standards.
Handle with the labels attached: the full framework text is gated, the exact phase naming inside it is not publicly documented, and the layer descriptions above are triangulated from the IAB's published summary and independent secondary coverage. The practical use is directional — it tells you where the IAB expects in-store measurement to converge: a verified baseline first, planning language second, deterministic outcomes last.
The privacy caution the standards compress into one line.
The December 2024 standards' entire privacy treatment is a note appended to the LTS definition: "NB: This might have GDPR implications." The sentence deserves more than a footnote, because everything above the reporting floor eventually touches it.
- Why LTS triggers it: likelihood-to-see is determined using sensor or analytic technology — cameras, computer vision, or equivalent instrumentation of shoppers. Whether the outputs are personal data depends on the design; whether the processing raises GDPR questions does not. The standard flags the issue and resolves nothing.
- It is not only LTS: Gross Impressions built on sensor-derived audience counts, dwell-time distance matrices, and one-to-one sales linkage all sit on the same instrumentation. The floor metrics can be built privacy-clean (door counters, transaction aggregates); the upper rungs get progressively harder.
- What to ask any sensor vendor: the legal basis for the processing, whether a DPIA exists, whether the approach is camera-less or anonymized by design, and where the raw sensor data lives and for how long. Secondary coverage of the 2025 maturity framework reports it addresses this with privacy-safe, anonymized approaches — verify against the gated full text before relying on it.
- The buyer's version of the question: if a network's OTS or LTS number depends on shopper instrumentation the network cannot explain, the number inherits the compliance risk. Put the privacy answer in the measurement spec, next to the formula disclosure the standards already require.
What to demand from an in-store proposal under these standards.
The standards are more prescriptive than most IAB documents — they name a reporting floor and require disclosure. That makes them directly usable as a checklist. Seven demands, each anchored in the text.
- 01
Name the rung
Ad Play, Gross Impression, OTS, and LTS are four different claims about the same playout. An "in-store impressions" line with no term attached is not reportable under the standards — carry the term with the number, always.
- 02
Show the formula
The standards require disclosure of the impression calculation and the traffic-collection method — transaction data, panels, sensors, door counters, or beacons. "Proprietary methodology" is not one of the permitted answers.
- 03
Map the exposure zone
Gross Impressions are defined against a Display Exposure Zone; audio defaults to the entire venue unless the retailer notes otherwise. Ask for the zone definition per placement and the dwell-time distance matrix the standards say retailers should provide.
- 04
Open the OTS multipliers
A benchmark OTS number is footfall times six or more modeled factors. Aisle penetration and touchpoint compliance are the levers — in the standards’ own worked example they cut the raw number by 36% before anything else moves. Ask for each input, sourced.
- 05
Match the control stores
The standards list the matching best practices: matched pre-period sales, controlled concurrent activity, identical price and promotion, same format and size. A lift study that cannot produce its matching table is a Sales Variance study wearing a lab coat.
- 06
Fix the windows before the flight
The standard reporting frame is 30 days pre-exposure, the campaign, and 30 days post — and retailers need to be able to provide flexible windows when asked. New-to-Brand claims need the purchase-cycle window stated: 0–6 weeks regular, 7–26 semi-regular, 27+ infrequent.
- 07
Ask the GDPR question on LTS
If a network or vendor sells likelihood-to-see, sensor instrumentation of shoppers is in scope. The 2024 standards flag it in one line and resolve nothing. Ask for the legal basis, the DPIA, and whether the approach is camera-less or anonymized by design — before the pilot, not after.
Frequently asked questions.
What are the IAB in-store retail media measurement standards?
A joint IAB and IAB Europe document — In-Store Retail Media: Definitions and Measurement Standards, final December 3, 2024 — that defines five store zones, a four-rung impression ladder (Ad Play, Gross Impression, Opportunity To See, LTS), exposure zones and traffic metrics, and sales-measurement rules for in-store retail media. It was built on the IAB/MRC Retail Media Measurement Guidelines and MRC Digital OOH guidelines, with roughly twenty retailers and RMNs participating, and currently scopes to digital screens, audio, and connected shopping formats.
What is the difference between Ad Play, Gross Impressions, OTS, and LTS?
Four progressively stronger claims about the same playout. Ad Play is a device log event — the ad rendered, no audience claim. Gross Impression counts individuals present in the Display Exposure Zone while the display is functional (audience times ad play), and the standards note it does not mean the ad was seen. Opportunity To See is the viewable-impression analogue — people who could pass by the activation. LTS adjusts for the likelihood people actually noticed the ad, using sensor or analytic technology. The terms are never interchangeable and never summable.
What should retail media networks report for in-store today?
Per the December 2024 standards: Ad Play and Gross Impressions, with the calculation formula and the traffic-collection method disclosed. Opportunity To See is positioned as the best available proxy for a viewable in-store impression on digital activations — the near-term aspiration — and LTS is the sensor-based frontier. The floor is deliberately modest; the disclosure requirement is what makes it comparable across networks.
What is the IAB Viable Framework for Maturing In-Store Media Measurement?
A December 9, 2025 IAB framework offering a standard measurement baseline retailers and vendors can adopt today and a staged path toward comparable in-store measurement — per the IAB’s published summary, structured as a baseline of verified impressions, an insights layer (reach, frequency, dwell time), and an outcomes layer preferring deterministic sales linkage. The full document is gated behind a free IAB account, and the exact phase naming is not publicly documented — verify against the full text before adopting its vocabulary. It is a framework, not an IAB/MRC standard.
How is in-store sales lift measured under the standards?
The recommended reporting frame is 30 days pre-exposure, the campaign period, and 30 days post-exposure, with flexible windows on request. Sales Lift covers the advertised product; Brand (Sales) Lift covers the total brand in the same category — equivalent to Halo in IAB Europe’s online standards — and both must exclude non-activated stores. For incrementality, the method hierarchy is randomized test-vs-control first, multivariate testing when externals cannot be controlled, and one-to-one linkage where technology and privacy allow. Simple before/during/after Sales Variance is explicitly demoted to a last resort.
Standing up in-store measurement — or auditing someone else’s?
The standards give you the vocabulary, the floor, and the disclosure rules. The operating work is turning them into a reporting spec: named rungs, sourced OTS inputs, matched control stores, fixed windows, and a privacy answer for anything sensor-based.