Retail Media

Retail media is the practice of a retailer monetising its own first-party audience and digital real estate by selling advertising — sponsored product listings, onsite banners and homepage takeovers, plus offsite ads (CTV, social, open web) targeted using the retailer's shopper data. A retailer's ad business is usually called a retail media network (RMN). The category sits at the intersection of Personalisation, Customer Data Platform (CDP) first-party data, and attribution — and is the advertising layer that, per run 102, makes Webrooming/ROPO spend measurable. This page records what sources report; it does not advise.

Firewall: every claim below is what a named source reports. See ../../CONTEXT.md Rule 1. This was a web-only run — the Reddit and YouTube practitioner streams were both unavailable, so the page carries no practitioner counter-narrative and is weighted toward analyst (IAB, ANA, eMarketer) and vendor (Skai, Osmos) sources. Vendor claims are flagged inline.

Why it matters in ecommerce

Retail media is the fastest-growing major ad channel and a high-margin profit pool for retailers, but its measurement foundations (especially incrementality) are contested. For an ecommerce product team it shows up as onsite ad surfaces competing with organic merchandising for the same pixels — a direct tension with Ecommerce Search & Discovery, Search Merchandising, and the Product Detail Page (PDP).

Market size & growth

Benchmarks (as-of 2026)

Market	Figure	Source
US retail media spend 2026	$69.33bn, up from $58.79bn (2025), +17.8% YoY	eMarketer, Retail Media Ad Spending Forecast H1 2026
US market concentration	Walmart + Amazon capture >89% of incremental US retail media dollars in 2026	eMarketer (H1 2026)
Europe retail media 2024	€13.7bn, +21.1% YoY, ≈one-fifth of European digital ad spend	IAB Europe 2025
Europe forecast	€16.9bn (2025) → €20.8bn (2026) → €24.9bn (2027) → €28.8bn (2028)	IAB Europe (via PPC Land)
Global digital retail media 2025	~$166bn, growing ~20% YoY (≈5× the 4.3% total-ad-market growth)	Osmos (vendor)
EU in-store format	70% of European retailers offer in-store digital screens; 54% of buyers investing	IAB Europe 2025

eMarketer reports retail media is outpacing both social and search ad-spend growth (as-of 2026). IAB Europe's buyers name first-party data activation and omnichannel integration as the top growth drivers (each cited by 39% of buyers, as-of 2025).

Onsite vs offsite

Sources distinguish two retail media surfaces (Osmos, vendor — definitional, low conflict-of-interest):

• Onsite — ads on the retailer's own site/app: sponsored products, category banners, homepage takeovers. Sits closest to the transaction, with high shopper intent and more direct measurement.
• Offsite — uses the retailer's audience data to target ads on CTV, the open web, social, audio and video, for awareness / consideration / retargeting.

Economics (as-of 2026)

• eMarketer (via Advertising Week) reports offsite retail media ad spend will grow 42.1% in 2025, nearly 3× the onsite growth rate.
• Osmos (vendor) cites onsite margins ~70–90%, offsite ~20–40%, and RMNs overall at 60–70% profit margins — far above traditional retail. Treat as illustrative, not audited.
• Osmos (vendor) frames RMN take-rate maturity as Year 1 launch 0.5–2% of total company revenue → Year 2–3 scaling 2–4% → Year 4+ mature 4–8%, with onsite CPCs ~$0.50–$3.00 and CPMs ~$10–$40 for mid-market retailers. Vendor framework, no primary benchmark — directional only.
• Skai (vendor) reports sponsored search drives >60% of US retail media spend today.

[!unverified] Walmart's ad revenue is cited at $6.4bn (FY2025), with ~one-third of Q4 FY26 operating income from advertising and membership (Osmos, vendor). The primary source would be Walmart's own filings (10-K / earnings release), which were not fetched this run — treat as unconfirmed.

Measurement & the incrementality debate

This is the contested core of retail media.

• IAB Europe states incrementality differs fundamentally from attribution and ROAS — those methods show what happened, not whether the marketing caused the result (as-of 2025).
• Skai (vendor survey) reports 71% of advertisers now rank incrementality their most important retail media KPI, but only 15% report strong measurement confidence while nearly half admit they are struggling (as-of 2025).
• Skai (vendor) found that when marketers were offered 10 definitions of incrementality, votes spread across them nearly equally — i.e. little industry agreement on how to even define it.
• eMarketer reports 61% of US retail business decision-makers use media mix modeling (MMM) to measure incrementality (as-of Dec 2025), making MMM the top of the incrementality stack.
• ANA's report "Retail Media Networks: Optimism Tempered with Caution" found 55% of marketers see lack of cross-platform standardization as the greatest challenge and 48% say there's insufficient data showing how RMN ads attribute to sales; marketers treat RMNs as a "have-to-buy" rather than a "want-to-buy" (as-of 2024-07).
• The ANA Media & Measurement Leadership Council is working with the Media Rating Council (MRC) on a list of measurement "must-haves and must-dos" for all RMNs (as-of 2025).

Criticism & open risks

• InternetRetailing reports practitioners argue incrementality will not be resolved in 2026 because there is no universal definition and each retailer defines/measures "new-to-brand" differently or not at all (as-of 2025–2026).
• Some practitioners warn retail media drives short-term sales to the detriment of long-term brand equity, and that many RMNs remain structurally disconnected from merchandising, marketing and store operations (InternetRetailing, 2025–2026).
• Skai (vendor) reports a prediction that if shoppers shift to researching via conversational/AI agents, "the value of those search ads will start to erode" — linking retail media's future to Agentic Commerce and AEO (Answer Engine Optimization) (as-of 2026).

Fashion / apparel angle

All apparel-specific claims this run came from a single vendor source (Osmos) with no primary benchmark behind them — low confidence, recorded as hypotheses:

• Apparel marketers allocate 70–75% of digital budgets to commerce media — claimed as the highest commerce-media share of any vertical (Osmos, vendor, as-of 2026).
• In apparel RMNs, low purchase frequency, higher basket values and strong seasonality are said to make editorial native and CTV outperform sponsored-product grids; sponsored products work for marketplace fashion (Amazon, Walmart) while native editorial outperforms in curated retailer environments; Target's Roundel is the most-cited apparel/beauty RMN proof point.
• 25% of respondents in Germany and Austria say they are likely/very likely to buy fashion directly through an AI assistant (Osmos citing a PwC/Strategy& survey; primary not fetched).

Key terms

Term	Meaning (as reported)
Retail Media Network (RMN)	A retailer's advertising business selling access to its audience and surfaces
Onsite	Ads on the retailer's own site/app — sponsored products, banners, takeovers
Offsite	Retailer audience data activated on CTV / open web / social / audio / video
Incrementality	Whether ad spend caused a sale, vs attribution/ROAS which show what happened
Take-rate	RMN ad revenue as a share of total company revenue
MMM	Media mix modeling — top reported method for measuring incrementality
New-to-brand	Buyer new to the advertiser; defined inconsistently across retailers

Frontier links

Retail Media Network (RMN) · Onsite vs Offsite Retail Media · Incrementality · Sponsored Products · Media Mix Modeling (MMM) · Roundel (Target RMN) · Walmart Connect · Amazon Ads · Criteo · Commerce Media · New-to-Brand (NTB) · Retail Media In-Store (Digital Screens) · Capture Rate