# Retail media networks as a profit engine
Amazon's advertising business generates tens of billions of dollars a year, and much of it comes from a simple idea: brands pay to appear at the exact moment a shopper is deciding what to buy. That revenue carries margins closer to a software company than a grocery store. In 2026, nearly every large retailer wants a piece of it.
This lesson explains what a retail media network is, why it prints money, and how to build one without wrecking the customer experiencecustomer experienceThe overall perception a customer forms of your brand across every interaction, from first touch to post-purchase support.Voir la définition complète →.
A retail media network (RMN) is a retailer's advertising business. The retailer sells ad space (on its website, app, in emails, on in-store screens) and targets those ads using its first-party data
Think of it as renting shelf space, except digital and far more measurable.
Three assets make it work:
That last point is the killer feature. Most advertising ends with a guess about whether it worked. Retail media closes the loop.
Amazon Ads is the model. When you search "coffee maker" on Amazon, the top results labeled "Sponsored" are paid placements. Brands bid for them. Amazon also sells display ads, video, and off-site placements, all powered by its shopper data.
Walmart Connect is Walmart's version, launched to monetize its massive store and online traffic. It combines online sponsored search with in-store assets like screens near the entrance and audio. Walmart's scale (hundreds of millions of weekly shoppers across physical stores) gives it something Amazon lacks: a huge offline footprint tied to loyalty data.
Other retailers, from Kroger to Target (Roundel) to Instacart, have built similar businesses. The IAB (Interactive Advertising Bureau) publishes free standards and research on retail media measurement worth bookmarking.
Selling ads on inventory you already own is close to pure profit.
The retailer already paid for the website, the app, the store, and the customer relationships. Adding an ad slot to a search results page costs almost nothing incrementally. When a brand pays for that slot, most of the payment falls to the bottom line.
Compare that to selling the product itself. Grocery net margins are famously thin, often low single digits. Ad revenue can run at margins several times higher. This is why a retailer's ad business can contribute a disproportionate share of operating profit even when it is a small share of total revenue.
For a CPG (consumer packaged goods) brand, the appeal is different: they are already spending to win the shelf. Now they can win the digital shelf too, with proof it worked.
Most retail media search ads use an auction. Advertisers bid to appear for specific search terms, and the highest effective bids win, subject to relevance.
Two common pricing models:
The core metric brands watch is ROAS (return on ad spend): revenue generated per dollar of ad spend. A ROASROASReturn on Ad Spend (ROAS) measures the revenue generated for every unit of currency spent on advertising, calculated as revenue divided by ad cost.Voir la définition complète → of 4 means four dollars of sales for every dollar spent. Because retail media closes the loop, ROASROASReturn on Ad Spend (ROAS) measures the revenue generated for every unit of currency spent on advertising, calculated as revenue divided by ad cost.Voir la définition complète → here is measured, not estimated.
Here is the logic of a second-price-style keyword auction in plain pseudocode:
for each ad slot on the search page:
eligible_ads = ads bidding on this keyword
for ad in eligible_ads:
rank_score = ad.bid * ad.relevance_score
winner = highest rank_score
winner pays just enough to beat the next competitorThe key lesson: the highest bid does not automatically win. Relevance matters, because showing junk ads drives shoppers away. This is the pricing engine and the customer-experience safeguard in one formula.
The data is the product. Here is how retailers turn it into revenue.
Audience segments. A retailer can let a brand target "households that bought a competitor's diapers in the last 90 days" or "lapsed buyers of our private-label coffee." These segmentssegmentsDividing a market into distinct groups of customers who share similar needs, characteristics or behaviours, so each group can be served with a tailored approach.Voir la définition complète → are built from purchase history, which is more predictive than the demographic guesses most ad platforms rely on.
Off-site extension. Retailers increasingly let brands use that shopper data to target ads on other platforms (social feeds, connected TV, the open web) then measure the sales back at the retailer's checkout. This expands the addressable ad budget well beyond the retailer's own pages.
In-store media. Screens, shelf displaysdisplaysThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.Voir la définition complète →, and audio inside physical stores are the newest frontier. Walmart, Kroger, and others are converting store traffic into measurable ad inventory. AttributionAttributionA framework for assigning credit to the touchpoints that contributed to a conversion, so you can measure which channels and interactions actually drive results.Voir la définition complète → is harder here, but the reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.Voir la définition complète → is enormous.
First-party dataFirst-party dataData collected directly from your own customers and prospects through your own channels: your most reliable and privacy-compliant source.Voir la définition complète → is valuable partly because third-party cookies (tracking files that followed users across unrelated websites) have been declining for years due to browser changes and regulation.
Regulations like the EU's GDPR (General Data Protection Regulation) and California's CCPA (California Consumer Privacy Act) govern how customer data can be used. The safe practice most retailers follow: aggregate and anonymize data so brands see segment-level insights, never individual shopper records. Do not treat this as legal advice; involve counsel before launching.
🎬 [VIDEO: "How Retail Media Networks Work" — youtube.com — a clear explainer on how retailers monetize shopper data and ad placements]
This is where retail media businesses live or die.
Ad load is the share of a page or session given over to paid placements. Push it too high and search results fill with sponsored junk, shoppers lose trust, conversion drops, and long-term loyalty erodes. The ad revenue looks great this quarter and the business decays quietly.
The tension is real: the ads team wants more inventory, the merchandising team wants the best product shown first, and the shopper wants to find what they came for fast.
Practical guardrails that mature RMNs use:
The best framing for leadership: retail media is a long-term relationship business, not a short-term yield-maximization game. The customer's attention is the scarce asset. Overharvest it and it disappears.
Vérification des acquis
1. Why is 'closed-loop measurement' described as the killer feature of a retail media network?
2. Why do retail media networks carry margins closer to a software company than to a grocery store?
3. A retailer wants to launch an RMN but worries about harming the shopping experience. What is the core tension it must manage?
4. Select ALL correct answers. Which assets make a retail media network work?
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers. What distinguishes a retailer's first-party data as the foundation of an RMN?
Sélectionnez toutes les réponses correctes.
If you are advising a retailer that wants to launch or grow an RMN, the rough order of operations looks like this.
1. Get the data house in order. Clean, unified customer and purchase data is the foundation. No data, no network.
2. Start with sponsored search. It is the highest-intent, highest-margin, easiest-to-measure format. Shoppers searching are already ready to buy.
3. Prove ROAS to brands. Give advertisers a self-serve dashboard showing sales driven. Trust in measurement is what pulls budgets from other channels.
4. Expand formats carefully. Add display, then off-site, then in-store, watching ad load and organic conversion at each step.
5. Standardize measurement. Adopt shared industry definitions so brands can compare your network to others. Fragmented metrics are the industry's biggest complaint.
Two trends matter. First, in-store retail media is scaling fast as retailers digitize physical shelves. Second, brands are pushing for standardized, comparable measurement across networks, because managing dozens of separate retailer dashboards has become a genuine operational headache. Retailers that make measurement easy will win share of ad budgets.