The $200 Billion Advertising Category Most Startups Are Ignoring
For most of digital advertising’s history, brand spend flowed to Google (search intent) and Meta (social targeting). The logic was simple: the two platforms had the most users, the best targeting data, and the largest sales forces. But a structural shift has been underway since 2022, driven by the deprecation of third-party cookies, Apple’s ATT changes, and the rise of a data asset that neither Google nor Meta has access to: purchase transaction data.
Retailers — and platforms with purchase or near-purchase transaction data — possess what advertising strategists now call the highest-quality targeting signal in digital media: what people actually bought, not just what they searched for or clicked on. Amazon built the first and largest retail media network on this insight, generating $60 billion in advertising revenue in 2025 from its ad platform alone. Amazon and Walmart together capture approximately 84% of US retail media budgets. But the category is rapidly expanding beyond the two incumbents.
The US retail media market is projected to reach $69.3 billion in 2026, up 17.9% year-over-year. Globally, retail media and commerce media combined exceed $200 billion. European retail media grew 22.1% in 2025 — more than triple the 6% growth rate for total ad spend. Carrefour, Tesco, and Instacart have each built meaningful retail media businesses. Over 80% of digital advertisers now allocate budgets to retail media, with three-quarters planning to increase those budgets in 2026.
The opportunity for startups is not to compete with Amazon — it is to replicate the model for the market segments, geographies, and customer types that Amazon and Walmart do not reach.
The Yassir-Kawarizmi Blueprint for Super-App Retail Media
The clearest emerging-market case study in startup retail media comes from Algeria’s Yassir, which in March 2026 acquired Kawarizmi, a Paris-based ad-tech and programmatic trading company, to accelerate its retail media network expansion across EMEA.
Yassir, founded in 2017 by Noureddine Tayebi and El Mahdi Yettou, operates a multi-service platform offering ridesharing, grocery delivery, and financial services to over 8 million users across 60+ cities in six countries. The company raised $50 million in Series B funding in 2022. Its first-party data asset — granular transaction records for 8 million users across rides, food delivery, and grocery purchases — represents exactly the kind of purchase-intent signal that brand advertisers pay premium rates to access.
Kawarizmi specializes in algorithmic media buying and data-driven advertising across Europe, Africa, the Middle East, and South Asia. By acquiring Kawarizmi rather than building ad-tech capabilities internally, Yassir acquired two things simultaneously: the technical infrastructure for programmatic ad serving and trading, and a EMEA-scale client roster of brands that Kawarizmi already serves. The acquisition collapses what would otherwise have been a multi-year build into a capability that can be deployed against Yassir’s user base immediately.
The model is replicable. Any platform with transaction-level first-party data — a delivery app, an e-commerce marketplace, a super-app with financial services, or a loyalty program operator — has the underlying data asset required to build a retail media network. The question is not whether the data exists, but whether the startup has the technical layer to monetize it and the brand relationships to fill ad inventory.
Advertisement
What Founders Building Retail Media Networks Should Do
1. Audit your first-party data for advertising-grade targeting signals
Not all user data is equally valuable to brand advertisers. The highest-value signal is verified purchase transactions — items bought, frequency, basket size, and category. The second tier is intent signals: product searches, cart additions, category browse behavior. The third tier is demographic and behavioral signals. Before investing in ad tech infrastructure, audit what data you actually have and map it against what advertisers will pay for. A platform with 500,000 users and verified purchase data in three product categories can generate more advertising revenue per user than a platform with 5 million users and only behavioral data. The advertising revenue formula is: (data quality × audience scale × inventory fill rate). Data quality multiplies everything else.
2. Build or buy the programmatic layer before approaching brands directly
The fastest path to retail media revenue is not a direct sales team calling brand managers — it is programmatic infrastructure that allows agencies and DSPs (Demand-Side Platforms) to buy your inventory through existing workflows. Brands allocate retail media budgets through agencies, and agencies buy programmatically. If your inventory is not accessible via a major DSP (The Trade Desk, DV360, Amazon DSP), you are invisible to the majority of brand budgets. The choice between building versus buying this infrastructure (as Yassir did with Kawarizmi) depends on your team’s technical capability and timeline. For most startups, the fastest path is an integration with an existing programmatic exchange rather than building a proprietary stack — full stack takes 18–24 months and $3–8 million in engineering investment.
3. Price on closed-loop attribution, not on CPM
The premium that retail media commands over traditional display advertising is its ability to prove that an ad impression caused a purchase. This closed-loop attribution — showing the advertiser which impressions led to verified transactions in your platform — is the pricing moat. Retail media networks that sell on CPM (cost per thousand impressions) compete directly with Google and Meta on the same metric where those platforms have a scale advantage. Networks that sell on CPA (cost per acquisition) or ROAS (return on ad spend) anchored to verified in-platform purchases command 2–4x the CPM equivalent because they are selling a guaranteed outcome, not a probability. Build your attribution model before you build your rate card.
4. Start with endemic advertisers, expand to non-endemic
Endemic advertisers are brands whose products are directly relevant to your platform’s purchase categories — a grocery delivery app selling ad inventory to food and beverage brands, for example. Non-endemic advertisers are brands buying for audience targeting rather than category adjacency — a financial services company advertising on a grocery app to reach high-income households. The proven retail media playbook is to launch with endemic advertisers (straightforward value proposition, faster sales cycles) and expand to non-endemic once attribution data demonstrates audience quality. Amazon moved in this sequence — it started with product ads for items sold on its platform before opening to brand campaigns by companies not selling on Amazon. Startups should map their endemic advertiser category and target the 10–20 brands most motivated to reach their specific customer base at launch.
The Structural Lesson
Retail media networks have 60–70% profit margins — significantly higher than the underlying commerce business they sit on top of. Amazon’s advertising segment generates higher operating margins than AWS. For startups with existing user bases and transaction data, this is a fundamental insight: the highest-margin business available to them may not be their core product, but the advertising layer they can build on top of it.
This reordering of margin priority is not unique to Amazon. It is the structural logic behind every major platform’s shift toward ad monetization — Meta, Google, Spotify, Netflix, and now super-apps across emerging markets. The advertising layer captures value that the core product creates but cannot directly monetize. For a delivery platform, each transaction creates a data record worth more to an advertiser than the margin on the delivery fee. For a fintech app, each payment creates a purchase signal worth more per impression to a brand than the interchange fee generates for the platform.
Founders who understand this structure position their user base as a media asset from the beginning — designing data collection, consent frameworks, and attribution infrastructure into the core product, rather than retrofitting it years later when the ad network ambition arrives. The Yassir-Kawarizmi acquisition works because Yassir had years of transaction data and user relationships already in place; Kawarizmi provided the ad tech layer to activate it. Startups that build both the data asset and the activation capability from early stages will not need an acquisition to unlock this margin tier.
Frequently Asked Questions
What is a retail media network and how is it different from traditional digital advertising?
A retail media network is an advertising platform built by a retailer or commerce platform that uses its own first-party purchase transaction data to target ads. Unlike traditional digital advertising (which uses cookies or behavioral proxies), retail media uses verified purchase data — what people actually bought — to prove that ad exposure caused a transaction. This closed-loop attribution is the key differentiator, allowing retail media networks to charge premium rates and demonstrate ROI directly to brand advertisers.
How did Yassir’s acquisition of Kawarizmi create a retail media capability?
Yassir acquired Kawarizmi, a Paris-based programmatic ad-tech company, in March 2026 to build a retail media network on top of its 8 million user base across six countries. Kawarizmi provides algorithmic media buying infrastructure, cross-market execution, and an existing EMEA brand client roster — the technical and commercial layer that Yassir needed to activate its first-party transaction data as advertising inventory. The acquisition gave Yassir an EMEA-scale retail media capability without the 18–24 month build timeline that a from-scratch ad tech infrastructure would require.
What profit margins can startups expect from retail media networks?
Retail media networks typically generate 60–70% profit margins on advertising revenue — substantially higher than the underlying commerce or logistics business. Amazon’s advertising segment is its highest-margin business unit, exceeding AWS margins. For startups, this means the advertising layer can generate more profit per dollar of user engagement than the core product, making it a strategically important revenue diversification that changes the overall economics of the platform.
—
Sources & Further Reading
- Algeria’s Yassir Acquires Kawarizmi to Accelerate Retail Media Expansion Across EMEA — Disrupt Africa
- Retail Media Market Outlook 2026: Key Data and Growth Forecast — Adtelligent
- Retail Media Trends 2026: What’s Driving a $203.9B Market — Rockbot
- Record-Breaking Funding in Q1 2026: AI Dominates Global VC — Crunchbase News
- How Top Retail Media Networks Are Monetizing 1st-Party Data — MadHive
