AI is already influencing the purchasing decisions of European consumers.

Artificial intelligence is already influencing how European consumers discover, compare, and choose products , even though fully autonomous purchasing by AI agents is still in its early stages.

This conclusion comes from McKinsey, in the article "Europe's agentic commerce moment: Decision influence is here; execution is coming," developed by QuantumBlack, the consultancy's AI unit.

According to the analysis, so-called agentic commerce should not be understood solely as the moment when a bot automatically buys on behalf of the consumer. In the short term, the most relevant impact is happening before the transaction: in the discovery, comparison, recommendation, and formation of the set of options that the consumer considers.

AI enters the funnel before the purchase.

McKinsey estimates that, by 2030, agentic commerce could orchestrate between three and five trillion dollars globally, as AI agents gain influence in discovery, decision-making, and transactions across various categories.

For brands and retailers, the strategic question is no longer just how to convert human customers. It also becomes how to remain visible, understandable, and recommendable when the first "customer" in the funnel is an artificial intelligence agent.

When evaluation shifts to AI-mediated interfaces, traditional factors such as positioning, advertising, or user experience remain relevant, but are no longer sufficient. Offers must now be retrievable, comparable, and defensible within the logic of an AI system.

84% use AI in their daily lives.

The McKinsey study is based on a survey of consumers in France, Germany, and the United Kingdom. According to the data, 84% of European consumers surveyed say they use AI tools in their daily lives. The most common uses include researching general topics, mentioned by 57%, and writing or improving content, indicated by 44%.

In the context of purchasing, 38% say they use AI tools to research products and services or to decide what to buy.

The use of AI in the purchasing journey is particularly evident in the initial stages. Among European consumers who use AI in the purchasing process, 63% use these tools to compare brands, models, prices, and reviews. Another 55% use AI to learn about a category or product, while 46% use these solutions to discover new products or gain inspiration.

These data show that AI is being adopted first as a decision support layer, streamlining research and comparison processes that previously required consulting multiple websites, platforms, and information sources.

Influence grows faster than execution.

McKinsey emphasizes that AI's influence on decision-making is growing faster than the infrastructure for execution or consumer confidence in complete autonomy. In other words, when a consumer arrives at a brand's or retailer's website , some of the evaluation work may already have been done within an AI interface.

This shift forces companies to think about product visibility beyond their own channels. Discovery can begin on ChatGPT, Gemini, agents integrated into retailers, or third-party shopping assistants.

European consumer confidence, in turn, varies depending on the role assigned to AI. Consumers are more comfortable when AI helps summarize reviews, compare options, highlight advantages and disadvantages, or recommend the best choice. In these cases, the technology supports judgment, but keeps the final decision in the hands of the person.

Trust diminishes when AI gets close to execution, such as filling shopping carts, completing checkout , or automatically placing recurring orders. McKinsey's conclusion is clear: consumers don't reject AI's judgment, but they resist its unchecked authority.

The delegation will have to be reversible and transparent.

For European consumers, delegating decisions to AI agents seems to depend on three conditions: reversibility, accountability, and explicit consent. Reversibility implies that the action can be undone. Accountability requires clarity about who is responsible when something goes wrong. And explicit consent presupposes well-defined limits on what the agent is authorized to do.

For companies, this means that trust should be treated as a product feature, not just a communication problem.

Adoption varies little between categories.

The use of AI to inform purchasing decisions is already relatively widespread. According to McKinsey, adoption is concentrated between 30% and 36% across various categories, suggesting that AI is functioning as a general evaluation layer and not just as a tool for highly technological purchases.

Usage tends to be slightly higher in more search-intensive categories, such as clothing and footwear, health and medical products, consumer electronics, and travel, where consumers face greater complexity, more alternatives, and a greater need for comparison.

One category stands out for its lowest adoption rate: optics. According to the study, only 13% of consumers use AI in this category. McKinsey's explanation is less related to a lack of informational complexity and more to the importance of physical fitting, in-person evaluation, and low purchase frequency.

This suggests that agentic commerce will not advance uniformly. Its development will depend on the category, the level of perceived risk, the need for physical validation, and the degree of consumer confidence.

Brand loyalty will be reconfigured.

McKinsey believes that mediation by AI agents does not necessarily mean the end of brand loyalty, but rather its reconfiguration. Loyalty may come to be expressed through explicit preferences given to the agents, such as authorized brands, excluded brands, or user-defined selection criteria.

At the same time, AI agents could present unfamiliar brands that better match the consumer's preferences, values, or aesthetics, broadening the range of considerations.

One of the main implications for brands and retailers is the need to compete for AI-mediated decisions, not just human attention. McKinsey speaks of the importance of machine readability. In practice, this means ensuring rich, structured, and consistent product data, clear taxonomies, coherent names, verifiable arguments, and credibility signals recognizable by automated systems.

Visibility no longer depends solely on advertising, placement , or visual experience. It also depends on a brand's ability to be understood, compared, and justified by an AI agent.

"Agent engine optimization" gains relevance.

The analysis points to the emergence of a logic close to what some are already calling agent engine optimization . Just as SEO forced brands and retailers to optimize content for search engines, agentic commerce may require companies to optimize products, claims , data, and reputation for AI agents.

In this new context, the priority is to ensure that offers are interpretable, recommendable, and defensible within automated comparison environments.

McKinsey identifies three priorities for companies that want to remain relevant in agentic commerce. The first is to compete for AI-mediated decisions, ensuring that products and brands can be found and understood by intelligent agents.

The second is to design AI experiences based on trust, with clear explanations, human confirmation, and the possibility of reversal.

The third is to prepare the execution infrastructure, namely payments, identity, authorization, and data integration, for a scenario in which AI-assisted transactions will gain greater relevance.

Conversational commerce may scale before full autonomy.

The consultancy also believes that conversational commerce may gain scale before fully autonomous delegation. As some AI environments begin to link recommendations to integrated purchase options, the boundary between discovery, comparison, and transaction becomes more fluid.

However, full autonomy should grow selectively, first in more limited, reversible, and lower-risk cases, before progressing to recurring decisions or those requiring greater involvement.

The study shows that AI is indeed altering the logic of the purchase journey. The immediate impact is not on the total replacement of the consumer, but on how preferences are formed and options are filtered before visiting a brand's website, store, or app.

For retailers and consumer brands, this creates a new competitive challenge: being present in the channels where the decision begins, even when those channels are not human, visual, or controlled by the brand itself.