Halo Attribution

Halo Attribution measures the sales lift that advertising generates beyond the directly advertised product — capturing uplift in related products, categories, or channels that last-touch attribution misses. The term is used in two distinct ways in the industry: a narrow IAB Europe standard definition for retail-media-network reporting, and a broader practitioner meaning covering cross-channel and cross-retailer spillover.

IAB Europe Standard Definition (V2.1, May 2026)

IAB Europe's Commerce Media Measurement Standards V2.1 (May 2026) define Halo Attribution as "Same Brand, Same Category (Category as defined by the CMN catalogue)" — that is, sales of any product in the same brand and same retailer-catalogue category as the advertised product, attributed to an ad exposure or click. (IAB Europe V2.1 PDF, 2026-05)

Key structural rules in V2.1 (as-of 2026-05):

Same SKU attribution is mandatory for sponsored product ads; Halo Attribution is optional but standardised when reported.
"Halo attribution can be provided for all formats." (IAB Europe V2.1, 2026-05)
Default lookback window: 30-day post-view and post-click, with flexible shorter windows permitted; Quick Commerce default is 7 days.
Future roadmap: "brands should be able to determine their own definition of Halo with the CMN in an automated way" — acknowledging the current "Same Brand, Same Category" scope is a baseline, not a ceiling. (IAB Europe V2.1, 2026-05)

IAB Europe V2 (the preceding version, announced January 22, 2026) added: a formal incrementality definition, gross/net sales distinctions, new-to-brand/new-to-category timeframe guidance, and quick commerce metrics. A six-month transition period ended July 2026, after which only V2/V2.1 certification is available. (IAB Europe press release, 2026-01-22)

The Two Meanings of "Halo"

The term "halo attribution" is used in materially different ways by the IAB standard and by industry practitioners — creating a definitional ambiguity that spans published research.

IAB Europe definition (narrow): uplift within the same brand and same category within a single retail-media-network (CMN). Scoped to on-RMN sales data. Does not cover cross-channel or cross-retailer halo. (IAB Europe V2.1, 2026-05)

Practitioner/industry definition (broad): any indirect sales lift that advertising generates beyond the direct attributed conversion, including:

Cross-channel halo: a Meta or TikTok campaign drives uplift on Amazon sales or brick-and-mortar — captured by Marketing Mix Modelling (MMM) or Conversion Lift holdout, but invisible in platform-native dashboards.
Cross-retailer halo: advertising on one RMN lifts organic sales on a competing RMN.
In-store halo: digital advertising (DSP, sponsored products) drives in-store purchases — invisible to online-only attribution.
Brand search halo: awareness spend (display, video) drives incremental branded search volume on Amazon or Google.

Both definitions appear in vendor content under the same label. Readers should verify which scope a source is using. (Osmos.ai, 2026-07-27)

How Halo Attribution Is Measured

Osmos.ai (2026) identifies four operational decisions that define any halo model — and notes that standardisation "is nearly impossible" due to differing product taxonomies across retailers (Osmos.ai, 2026-07-27):

Credit exact product advertised only (Same SKU — IAB mandatory baseline)
Same product in a different size/variant (Same Parent SKU)
Same-category SKUs of the same brand (IAB Halo standard)
Cross-category products of the same brand (beyond IAB scope; Criteo's "Same Brand" level)

Measurement methodologies used in practice (fusepoint / Power Digital, 2025-09-25, updated 2026-03-30; fusepointinsights.com):

Matched Market Testing (MMT): select paired test/control markets, vary retail media spend in test only, measure total sales including DTC and retail partners. Reveals cross-channel and in-store halo directly.
Marketing Mix Modelling (MMM): uses 2–3 years of blended sales and spend data, controlling for promotions and seasonality. Best for always-on halo visibility across channels.
Incrementality holdout (Geo Holdout Testing): withholds ads from a control group to isolate causal lift.
Data clean rooms: IAB Europe V2.1 approves three delivery methods: (1) CMN-provided reporting, (2) open data/API for brand-side analysis, (3) data clean rooms. Each has different cost, interoperability, and flexibility trade-offs.

Platform Implementation: Criteo vs IAB Standard

Criteo Commerce Max uses a three-level product-matching hierarchy that implements a similar construct to IAB Halo but with a different scope:

Criteo Level	Scope	IAB V2.1 Equivalent
Same Product	Exact SKU matched	Same SKU (mandatory)
Same Category	Same category as advertised product	IAB Halo (same brand + same category)
Same Brand	Any product from the same brand, across all categories	Beyond IAB scope

Criteo does not use the term "halo attribution" — it uses "same brand" / "same category" product matching terminology. (Criteo Commerce Max Help Center, undated)

IAB Europe V2.1 "Halo" (V2.1 PDF, 2026-05) restricts Halo to "Same Brand, Same Category" simultaneously — a combined two-axis scope. Criteo's "Same Brand" level (Criteo Commerce Max Help Center, undated) extends to all categories for the same brand — an explicitly cross-category scope excluded from the current IAB standard. This is a definitional tension between the standard and at least one major platform's implementation.

Benchmarks (as-of 2026-08-10)

The Criteo +428% incremental ROI figure is from a 2025 US-only study (44,000+ campaigns, Aug 2023–Oct 2024); no equivalent EMEA figure was confirmed.

Cross-channel and cross-retailer halo:

Fospha (ClickZ, 2026-04-21): across hundreds of retail brands, 40% of Amazon revenue is influenced by non-Amazon paid media (primarily Meta and TikTok) — entirely invisible to platform-native attribution tools. For beauty and apparel brands with a high Amazon index, this figure "often exceeds 40 percent." (clickz.com) (as-of 2026-04-21)
Fospha — TikTok ROAS correction (2026-04-21): when Amazon halo is included in measurement, TikTok's reported ROAS moves from 2.1 to 2.8 — approximately +33% uplift on the measured ROAS figure. (clickz.com) (as-of 2026-04-21)
Incremental ($350M spend, 150,000 campaigns, 2026): siloed last-touch attribution misses up to 80% of offsite video's real impact once cross-retailer halo and in-store sales are accounted for. 40%+ of DSP-driven CPG sales happen in-store, not online. (incremental.com) (as-of 2026; publication date unconfirmed)

Two "40%" figures appear in the literature for different phenomena: Fospha (2026) reports 40% of Amazon revenue influenced by non-Amazon paid media (clickz.com). Incremental (2026) reports 40%+ of CPG DSP-driven sales happen in-store (incremental.com). These measure different effects (cross-channel paid media → Amazon; vs Amazon DSP → in-store) and should not be conflated.

Measurement maturity:

Skai (2026 State of Retail Media, n=166 advertisers, partnered with Stratably; 2026-02-04): only 15% of marketers say their organisation is very or extremely effective at measuring retail media performance overall. 75% cite incrementality as the top measurement challenge; 59% cite cross-channel measurement. Only 20% are both good at measuring and applying insights to decisions. (skai.io) (as-of 2026-02-04)
EMEA RMN fragmentation (IAB Europe, 2025-04-30): EMEA has over 138 retail media networks, but most advertisers use only 3–5; "no easy way to buy or measure consistently across them" — making cross-RMN halo deduplication structurally difficult. (IAB Europe YouTube, 2025-04-30) (as-of 2025-04-30)

The EMEA 138-RMN figure is from an April 2025 IAB Europe session — actual count may have grown by August 2026.

Fashion and Apparel

No study was found quantifying fashion/apparel intra-category halo lift at the product taxonomy level (e.g. advertising a jacket → lifting accessory or footwear sales within the same brand). All fashion-specific data relates to cross-channel halo (paid social → Amazon).

Fospha (ClickZ, 2026-04-21): for beauty and apparel brands with high Amazon index, cross-channel halo (paid social → Amazon) "often exceeds 40 percent" of Amazon revenue — proportionally larger than CPG. (clickz.com)
Necessaire case study (via Fospha/ClickZ, 2026-04-21): using halo-inclusive unified DTC + Amazon ROAS model to justify sustained upper-funnel investment, the brand achieved 47% higher Prime Day revenue than industry benchmarks. (clickz.com) (as-of 2026-04-21; vendor case study — single source)
DTC skincare brand — MMM halo reveal (fusepoint, 2025-09-25 / updated 2026-03-30): Meta prospecting paused for weak ROAS had been generating +18% lift in Sephora sell-through; pausing it caused Sephora sales to fall. Only MMM revealed the connection.

The fusepoint DTC skincare case study (2025-09-25) and Necessaire case study are pre-2026 or unverified date; both are vendor-adjacent sources.

Industry Adoption and Standardisation Challenges

The measurement problem: Jason Wescott (WPP Media, Chair of IAB Europe's Retail & Commerce Media Committee), January 2026: "our research shows that a lack of standardisation remains the biggest barrier to Retail and Commerce Media growth." (IAB Europe, 2026-01-22)

Osmos.ai (2026) identifies three structural failure modes that suppress halo visibility in practice:

Platform silos: each RMN reports in its own dashboard with its own attribution definitions.
Attribution inflation: multiple platforms claim the same conversion.
Cross-channel blindness: retail media lifts DTC and vice versa, undetected without incrementality methodology.

The ROAS vs incrementality debate (IAB Europe Great Debate, 2025-04-30): Marta Piotrowska (Allegro) argued ROAS is "built on flawed click-based attribution models rooted in the search-dominated world of 2010" and does not capture halo effects. Paul Dahill (Koddi) countered that incrementality "is still complex, often inconsistent, and not scalable for all retailers, particularly smaller ones," warning against a "measurement arms race."

Incrementality as the future standard: Piotrowska (Allegro) and Wescott (WPP Media/Skai 2026) frame incrementality as the natural successor to ROAS for capturing halo and brand-building effects (IAB Europe YouTube, 2025-04-30; Skai 2026-02-04). Incrementality as inaccessible: Dahill (Koddi) and Yara Daher (IAB Europe, 2025-03-06) argue it remains too complex and non-scalable, particularly for smaller RMNs; Daher writes that incrementality "is one piece of the measurement funnel, not the entire solution." Audience polling at the Great Debate shifted from 65% (incrementality will be king) to 52% (same view) by session end, indicating practitioner scepticism.

IAB Europe plans to extend standards beyond retail and quick commerce to travel and finance media networks in 2026. (IAB Europe, 2026-01-22)

Key terms

Term	Meaning
CMN	Commerce Media Network — IAB Europe's term for retail media networks
Same SKU	Exact product (or parent SKU) purchased post-ad exposure — IAB mandatory baseline
Halo Attribution	IAB V2.1 optional layer: same brand + same category purchases beyond Same SKU
Cross-channel halo	Practitioner term: ad on one platform (Meta/TikTok) lifts sales on another (Amazon/store)
Cross-retailer halo	Sales lift on a different RMN from advertising on the primary RMN
In-store halo	DSP or sponsored product advertising lifts physical store purchases
Matched Market Testing (MMT)	Geo-based experiment to isolate halo across channels/retailers
Data clean room	Privacy-safe environment for brands and CMNs to join first-party datasets for measurement

What practitioners report

Michael Kaminsky (Recast co-CEO, Total Retail, 2025-11-11): brand search is the most common halo mechanism; a TikTok awareness campaign may drive modest DTC sales while contributing incremental lift to Amazon and brick-and-mortar, but last-touch tools credit search or Amazon — not TikTok.
Colin Lewis (Retail Media Specialist, Retail Media Pulse 2026, 2026-06-04): the halo measurement problem is a "data-to-knowledge translation failure, not a data absence problem."

Michael Kaminsky quote is from November 2025.

Gaps

No IAB Europe V2 vs V2.1 changelog (what precisely changed between V2 Jan 2026 and V2.1 May 2026) was found.
No fashion/apparel intra-category halo benchmarks (same brand, same category at the SKU level) were found; published lift studies exist almost exclusively in CPG.
No Roundel (Target) specific halo attribution benchmarks found.
No EMEA-specific Criteo halo benchmarks confirmed from public primary sources (2024 data existed in vault prior to this run).
Reddit MCP unavailable (known Cowork cloud gap) — practitioner debate on halo methodology absent.
YouTube transcripts unavailable (Apify MCP gap) — all video findings are from wrap-up pages and the written standard, not direct transcripts.