Commerce Media Network (CMN)

Table of Contents

  1. Definition
    1. How It Relates to Marketing
  2. How Commerce Media Networks Work
  3. How to Utilize Commerce Media Networks
  4. Comparison to Similar Approaches
  5. Best Practices
  6. Future Trends
  7. Frequently Asked Questions
  8. Related Terms
  9. Sources

Definition

A commerce media network (CMN) is an advertising business built on a company’s first-party purchase data and owned audience touchpoints, letting brands reach shoppers at or near the moment of buying. It generalizes the retail media network — the model pioneered by Amazon and Walmart, where a retailer monetizes its own shopper data and on-site inventory — to any business that owns transaction data and customer attention. Marketplaces, travel platforms, financial services, delivery and rideshare apps, and grocers can all run one. “Commerce media” is the umbrella category; “retail media” is its largest and most mature subset. The terms get used interchangeably, but commerce media is the broader frame, reflecting how the model has spread beyond traditional retailers and off the retailer’s own site.

What makes the model distinct is its position at the intersection of commerce and media. A retailer or platform possesses data most advertisers can’t get anywhere else — actual purchases, basket contents, browsing, and intent — and pairs it with places to show ads. That combination is why retail media stopped being a lower-funnel line item and moved closer to the center of growth. As of 2026, the category is also being reshaped by AI agents, which is where most of the strategic attention now sits.

Back to Top

How It Relates to Marketing

Commerce media matters to marketers because it’s where high-intent shopper data meets ad inventory, and increasingly where AI is rewiring both. The convergence with agentic commerce is the live story. Retail media was built on a simple premise: influence the customer before they press checkout. Agentic commerce challenges that premise directly, because if software is choosing products, comparing alternatives, and completing purchases, the impressions, clicks, and dwell time a commerce media network monetizes and measures start to disappear.

One framing of the problem is a visibility gap. In traditional ecommerce a retailer sees everything — impressions, clicks, add-to-cart, drop-offs. When discovery and consideration move inside ChatGPT or Gemini, that activity happens somewhere the retailer can’t observe, so attribution collapses, personalization breaks, and, as one analysis put it, retail media goes dark. The demand is real but the infrastructure lags: AI-generated recommendations have shown roughly 4.4x higher conversion than traditional search in McKinsey’s work, yet agent-driven conversion has trailed affiliate channels because most merchant systems weren’t built for agents. The opportunity sits on the other side of that gap. Rather than treating agents as a threat to ad inventory, the networks expanding into agentic touchpoints stand to open new surfaces for discovery and monetization — agentic commerce becomes, as Mars United describes it, another point of entry to browse a different formulation of the retailer’s shelf.

Back to Top

How Commerce Media Networks Work

The mechanics rest on three assets: first-party data, owned media, and measurement. The network uses its purchase and behavioral data to build audience segments, sells access to those audiences and its ad placements to brands, and closes the loop by tying ad exposure back to sales. That closed-loop measurement — proving an ad drove a purchase using the network’s own transaction data — is the model’s core advantage over ad networks that can only infer outcomes.

Inventory has expanded well past the on-site search bar. Networks now run sponsored placements on their sites and apps, in physical stores, and increasingly off-site through open exchanges, preferred DSPs, social, and connected TV. Walmart’s 2024 partnership with Disney is a clean example of the off-site direction: advertisers use Walmart’s audience segments to buy ads on Disney’s streaming services, with sales impact measured through clean rooms. Data collaboration tools like clean rooms are what make that cross-company matching possible without exposing raw customer data.

The model isn’t without friction. Fragmentation is a real complaint — eMarketer found 55% of US advertisers reporting inconsistent targeting and attribution across retail media networks, which is one reason 2026 buyers want unified dashboards and the ability to plan commerce media alongside other channels rather than network by network.

Back to Top

How to Utilize Commerce Media Networks

For advertisers, a commerce media network is a way to reach shoppers with intent the network can verify and to measure impact against real sales. The practical work is selecting networks with the audience and inventory that fit, planning commerce media within a broader channel mix rather than in isolation, and pushing for consistent measurement across the networks in play. The off-site and CTV expansion means brands can now use a retailer’s shopper data to target and measure beyond the retailer’s own properties.

For retailers and platforms operating a network, the agentic shift sets the near-term agenda. The recurring guidance across 2026 analyses points the same direction: optimize product data so it’s consistent, structured, and agent-readable across every touchpoint; establish real-time synchronization of inventory, dynamic pricing, and promotions; and build partnerships with LLM platforms to enable affiliate integration and better product recommendation. Many retailers are also deploying their own on-site AI agents to keep discovery and monetization within their walls. The strategic logic is that retailers remain the core of commerce through their control of purchase data, inventory, promotions, pricing, basket building, fulfillment, and payment — so the move is to extend the network into agentic surfaces rather than cede them.

Adoption is early, which is itself the opportunity. Salesforce’s Connected Shoppers Report found only two in five retail media networks piloting agentic commerce, even as a large majority of shoppers have folded LLMs into their buying journey. The gap between consumer behavior and network readiness is where competitive advantage is being decided.

Back to Top

Comparison to Similar Approaches

ModelData sourceInventoryDistinguishing trait
Commerce Media NetworkFirst-party purchase data from any commerce businessOn-site, in-store, off-site, CTVBroad category spanning retailers and non-retail platforms
Retail Media Network (RMN)A retailer’s first-party shopper dataThe retailer’s owned properties, plus off-siteThe largest, most mature subset of commerce media
Traditional Ad NetworkThird-party and inferred dataPublisher inventory across the webNo direct purchase data or closed-loop sales proof
MarketplaceTransaction and seller dataThe marketplace’s own surfacesFacilitates sales; advertising is secondary

The relationship to retail media is the one to keep straight: a retail media network is a commerce media network run by a retailer, and commerce media is the wider category that also covers travel, fintech, delivery, and other transaction-data owners. The sharper contrast is with a traditional ad network, which targets on third-party or inferred signals and can’t prove a sale the way a commerce media network can with its own purchase data. That closed loop is the whole point.

Back to Top

Best Practices

  • Plan commerce media within a unified channel strategy rather than network by network, and push for consistent targeting and attribution to counter fragmentation.
  • Prioritize closed-loop measurement, the model’s core advantage, and tie ad exposure to actual sales.
  • For operators, make product data consistent, structured, and agent-readable across every touchpoint as a precondition for agentic monetization.
  • Establish real-time synchronization of inventory, pricing, and promotions so agent-facing surfaces stay accurate.
  • Build LLM-platform partnerships for affiliate integration and product recommendation rather than waiting for agents to route around you.
  • Use clean rooms for off-site and cross-company data collaboration without exposing raw customer data.
  • Track the visibility gap directly by measuring AI referral traffic and agent-driven conversion alongside traditional on-site metrics.
Back to Top

Two forces are converging on the category. Off-site and CTV expansion continues to pull commerce media beyond owned properties, with more retailers connecting first-party data to open exchanges and DSPs and unifying it with other channels. At the same time, agentic commerce is forcing a rethink of where and how attention gets monetized. The honest state of play is unsettled — the industry doesn’t even share a single definition of agentic commerce, with some practitioners reserving the term for fully autonomous, no-human-approval purchasing and treating everything else as an evolution of search.

The forcing mechanism over the next two years is measurement. As boards demand proof rather than experimentation, networks will live or die on whether they can show outcomes in a world where much of the journey happens inside an assistant they can’t see. Payments and authorization infrastructure built for machine-driven transactions is part of that buildout. For the broader picture, the commerce media network is where the money side of agentic commerce concentrates: it’s the existing advertising machine that has to adapt to agent-mediated discovery, and the networks that extend into agentic touchpoints rather than defend the old funnel are the ones positioned to keep monetizing attention as it moves.

Back to Top

Frequently Asked Questions

1. What is a commerce media network? An advertising business built on a company’s first-party purchase data and owned audience touchpoints, letting brands reach shoppers near the moment of buying. It generalizes retail media to any business with transaction data, including non-retailers.

2. How is it different from a retail media network? A retail media network is a commerce media network operated by a retailer. Commerce media is the broader umbrella that also includes marketplaces, travel, financial services, and delivery platforms. The terms are often used interchangeably.

3. What makes the model valuable? First-party purchase data plus owned inventory plus closed-loop measurement. The network can target on verified intent and prove an ad drove a sale using its own transaction data, which traditional ad networks can’t.

4. How is agentic commerce affecting commerce media? It’s creating a visibility gap. When discovery and consideration move into AI assistants, the impressions and clicks networks monetize and measure happen out of view, so attribution and personalization break down — while new agentic surfaces open up for monetization.

5. Are retailers losing control to AI agents? Not necessarily. Retailers still control purchase data, inventory, pricing, fulfillment, and payment, so agentic commerce becomes another entry point to their shelf. The risk is to networks that don’t extend into agentic touchpoints.

6. How ready is the industry? Early. Salesforce found only about two in five retail media networks piloting agentic commerce, even though most shoppers already use LLMs in their buying journey. That gap is where advantage is being won.

7. What should operators do first? Make product data structured and agent-readable, synchronize inventory and pricing in real time, and build LLM-platform partnerships for recommendation and affiliate integration.

8. What about off-site and CTV? Commerce media has expanded well beyond owned sites into open exchanges, social, and connected TV. Walmart’s partnership with Disney, using Walmart audience data to target streaming ads measured via clean rooms, is a representative example.

Back to Top
Back to Top

Sources

Back to Top

Tags: ,

Was this helpful?

Previous

Agentic Commerce Protocol (ACP)

Next

Product Feed Optimization for AI