Online commerce relies on sales channels, advertising formats, and regulatory frameworks that evolve every quarter. Understanding these mechanisms allows for a trade-off between available growth levers and anticipating legal constraints that redefine the rules of the game for merchants, publishers, and brands.
Retail media: an advertising channel that reallocates marketing budgets
Retail media refers to the purchase of advertising space directly on online sales platforms (marketplaces, distributor sites). The advertiser targets consumers at the precise moment they are comparing products, which shortens the journey between ad exposure and purchase.
The global retail media market has surpassed $200 billion. Amazon alone captures an overwhelming share of this spending: in the United States, the platform accounted for 78% of retail media investments in 2025. In Europe, in key markets (France, Germany, Italy, Spain, United Kingdom), more than two-thirds of retail media budgets are also absorbed by Amazon.
This concentration creates a risk of dependency for brands that allocate an increasing share of their sales to it. French distributors are developing their own media agencies (Carrefour Links, Valiuz at Auchan), but their audience remains significantly smaller. To find all the news on blogbusiness.fr, several analyses detail the possible trade-offs between these platforms.
Generative AI and editorial content: a still unclear compensation model
Search engines integrate AI-generated responses directly into their results pages. For an online content publisher, the question is no longer just about appearing in the results, but whether their content feeds these responses without generating traffic in return.
Google is considering a compensation mechanism for publishers whose content is used by its generative AI systems. This model differs from neighboring rights in the press: compensation would be indexed to the value attributed by Google, not on the volume of use. The publisher controls neither the scale nor the selection criteria.
For businesses whose economic model relies on organic traffic (professional blogs, specialized media, comparison sites), this mechanism introduces a new form of dependency. Visibility in traditional results decreases when an AI response directly satisfies the user’s query, and the proposed financial compensation remains opaque.
Concrete consequences for a content strategy
- High-value informational articles (guides, analyses, exclusive data) are the most likely to be absorbed by AI responses. Protecting this content with paid access or limiting the indexing of certain pages becomes a strategic trade-off.
- Conversational and community formats (forums, user-generated FAQs) fare better, as their value lies in the diversity of viewpoints rather than in a single summary.
- Tracking the click-through rate from results pages (organic CTR) becomes a more reliable indicator than average position for measuring the real impact of SEO.
European regulations DMA and DSA: operational constraints for online sellers
The Digital Markets Act (DMA) and the Digital Services Act (DSA) are two European regulations that govern the practices of large platforms and the responsibilities of digital intermediaries, respectively. They are already in effect, and their effects are becoming clearer quarter after quarter.
The DMA targets “gatekeepers”: search engines, marketplaces, mobile operating systems. It imposes interoperability, prohibits self-preference in results, and opens the possibility for third-party sellers to contact their customers outside the platform.
What the DSA changes for merchants
The DSA imposes transparency obligations on targeted advertising and strengthens the fight against counterfeit products. Every online seller must be able to justify the traceability of their products and respond to information requests within set deadlines.
For small businesses, compliance represents a real organizational cost: updating general terms of sale, establishing a single point of contact, documenting customer review moderation processes. Large marketplaces have begun requiring their third-party sellers to provide proof of compliance, which gradually filters out the least structured operators.
Agentic commerce: when AI buys on behalf of the consumer
Agentic commerce refers to a model in which an artificial intelligence agent performs purchasing tasks on behalf of a user: price comparison, product selection based on predefined criteria, automated order placement.
This model alters the decision chain. The consumer no longer browses product sheets or results pages. The agent filters, ranks, and purchases based on technical parameters (price, delivery time, average rating, compatibility). Traditional marketing levers (visual identity, brand storytelling, packaging) lose some of their influence when the interlocutor is an algorithm.
For online sellers, the focus shifts to the quality of structured data associated with products: standardized descriptions, complete technical attributes, data feeds compatible with agent standards. A poorly structured product catalog becomes invisible to purchasing agents, even if the sheet is correctly indexed for a human reader.
The first applications concern recurring purchases (office supplies, consumables, cleaning products) and markets where comparison is based on objective criteria. Sectors where the decision involves subjective judgment (fashion, accessories, home decoration) remain less affected for now, but expansion is predictable as agents integrate stylistic preferences.
The interplay between retail media, the European regulatory framework, and the rise of agentic commerce is reshaping priorities for companies selling online. Mastery of product data, compliance with new obligations, and diversification of advertising channels are no longer tactical options but operational prerequisites.
