The era of purchasing agents: a turning point for specialist trade

Long perceived as a distant technology or reserved for digital giants, artificial intelligence has now entered a phase of maturity that directly concerns specialized retail. Garden centers and pet stores are no exception.

Since 2022, generative AI has demonstrated its ability to produce content: texts, images, videos, product sheets, customer responses or marketing materials.

But a new stage is being reached: that of agentic AI.

Unlike AI used for simple, ad-hoc instructions, agentic AI operates with a specific objective in mind. It is capable of planning actions, mobilizing tools, drawing on memory (data, documents, history), and acting semi-autonomously within a defined framework.
We are no longer simply talking about tools, but about new ways of orchestrating work and decision-making.

These are already widespread uses among consumers.

This shift is not theoretical. In the United States, Nearly 25% of consumers report having already made a purchase via a conversational interface or an AI agent..
Search, comparison, recommendation, basket creation, choice of delivery or collection options: the agent becomes a full-fledged intermediary in the purchasing process.

This evolution marks a profound break with traditional journeys based on keyword searches. Consumers no longer navigate from site to site: they express an intention, and AI takes care of it.

From SEO to GEO: a strategic challenge for retailers

In this new context, The challenge for brands is no longer simply having a well-written and SEO-optimized website..

Tomorrow, the real issue will be to be present in AI modelsand above all to make its products, data and expertise available consultable, understandable and usable by agents.

Specifically, these are the agents who:

will consult the product catalogues,
will analyze the features, compatibilities and reviews,
will check availability and prices,
will directly offer coherent baskets to customers.

The strategic question then becomes: Are our products and data identified as reliable sources by AI agents?

Being visible will no longer be enough. You will have to be Understood, credible and recommendedThis is the whole point of the transition from SEO to... GEO (Generative Engine Optimization).

Very concrete impacts on trade

Agentic AI is not limited to theoretical promises. Several players in commerce and services are already using it to profoundly transform their organization, processes, and customer relationships.

At Walmart, agentic AI is deployed across the entire value chain.

On the customer side, conversational shopping assistants are capable of suggesting coherent shopping carts, comparing products, summarizing reviews, and proposing compatible alternatives, even enabling direct purchase without going through a traditional e-commerce website.

Internally, agents analyze sales, explain underperformance, adjust forecasts and trigger restocking or assortment management actions under human supervision.

In the services sector, Booking has integrated agents capable of guiding the user from start to finish: formulating their intent, comparing offers, personalized recommendations, adjusting criteria, and finalizing the booking using natural language.

The agent no longer simply displays results: he guides the decision and radically simplifies the process.

These logics are now directly accessible in environments like ChatGPTwhere agents can:

to analyze a complex request expressed in natural language,
compare offers, products or services,
to synthesize heterogeneous opinions and data,
propose a ready-to-use decision or action plan.

For companies in the specialized retail sector, the applications are immediate:

internal assistants capable of answering a large proportion of after-sales service and field-related questions.
agents who assist in the development of product ranges and product sheets,
Automated sales and inventory analysis with actionable recommendations,
decision-making support for purchasing, marketing or supply chain teams.

Agentic AI, therefore, does not simply optimize what already exists. It reshapes the way organizations decide, arbitrate, and act, by moving from a manual execution logic to an intelligent orchestration logic.

Giving teams back time and meaning

One of the major contributions of agentic AI lies not only in productivity. It also relates to the recomposition of the human role.

By automating repetitive, time-consuming or low value-added tasks, AI allows teams to refocus on what makes specialized commerce strong: advice, expertise, education, relationship and trust.

For a sector based on living things, proximity and customer support, this issue is central.

A framework to be built collectively

These transformations are accompanied by essential questions: data governance, transparency of uses, cybersecurity, regulatory compliance, particularly in the context of the European AI Act.

AI cannot be a strategic blind spot. It must be part of a clear, shared and responsible framework, with a mapping of uses, rules for human supervision and a progressive skills development of teams.

Anticipate today to remain in control of the game tomorrow

Agentic AI therefore poses not only a technological question, but also a question of strategic positioning.

Being present in the models, making one's products understandable to agents, structuring one's data and processes is becoming a new prerequisite for commerce.

For garden centers and pet shops, the stakes are clear: not to let other actors speak for themto recommend on their behalf, to decide on their behalf.

By investing in these new standards today, specialist retailers can remain in control of customer relations, while gaining in efficiency and clarity.

Agentic AI will not make garden centers disappear. It will simply decide which ones will remain visible.

Thomas Le Rudulier - General Delegate - Garden Centers & Pet Shops of France

See all our news