Artificial intelligence is profoundly redefining the way e-commerce is used. With the emergence of what is known asAgentic Commerce, it is no longer just consumers who make their purchases, but AI agents who take direct control of the entire purchasing journey, from research to order. This paradigm shift is turning brands and retailers on their head: how can they exist in a customer journey that is increasingly automated, with human intervention only at the point of receiving the parcel?
Two experts in e-commerce and logistics performance met to discuss this central question: Loïc Lejal, Associate General Manager at Antadisan independent digital agency specializing in strategic and technical support for online merchants, and Christian Lelaidier, Indirect Sales Director at Shippingbo.
- LLM & agentic commerce: how does it work?
- Delivery, a key experience lever
- Capitalize on the post-purchase experience
Both have combined their expertise (strategic consulting, customer experience, digital architecture on the one hand, and mastery of logistics flows on the other) to offer a concrete vision of the challenges ahead. While AIs are becoming the new decision-makers, delivery remains the first point of physical contact with the customer.
In this session, speakers explore the new expectations imposed by AI agents, the levers for transforming logistics into a competitive advantage, and the impacts on customer loyalty, reputation and satisfaction. A key event for understanding how to prepare your e-commerce for the post-navigation era.
LLM & agentic commerce: how does it work?
Agentic commerce marks a turning point in e-commerce practices. It leverages the capabilities of large-scale language models (LLMs), such as ChatGPT or Gemini, to transform a simple intention into an act of purchase… without human intervention. AI becomes the new intermediary, capable of searching, comparing, selecting and ordering, in a matter of seconds.
A new way of formulating requests
The most visible change is in the way you search for a product. Where conventional search engines require a few keywords, AI expects a rich, nuanced, contextual prompt. We no longer type “best cervical pillow”, but “Which pillow is best suited to support the nape of the neck when you sleep on your side, without waking up in pain?”.
This shift towards conversational search is changing the way products are discovered. The conversational agent analyzes the intention behind each request and proposes precise solutions, adapted to the user’s situation. More than a search engine, it’s an answer engine.
An AI-driven purchasing process
Once the request has been analyzed, the AI takes control of the entire process. It first selects the most relevant product, then identifies the merchant who can deliver it under the best conditions. This double filtering is based on several criteria, such as availability, price, perceived quality, reviews, merchant status and payment terms.
At this stage, the user no longer interacts with a website, but remains in his messaging interface. They can then validate the purchase via an integrated checkout, without having to enter their details manually. The AI automatically completes the necessary fields, then triggers payment via Stripe, Apple Pay or Google Pay.
Here’s how the new route is organized:
- User request in the form of a prompt (average 30 words)
- Recommendation of one or more products
- Automatic merchant selection
- Add product to basket (currently in single-product format)
- Checkout auto-completion
- Integrated payment and order confirmation
This process is already underway at selected Shopify merchants and on the Etsy platform in the USA. The arrival of these features in Europe is imminent.
What AI expects from a merchant
In this model, logistics optimization becomes a visibility criterion. To rise in the suggestions of ChatGPT or Google AI Mode, a merchant must prove that it is reliable, fast and well rated. It’s no longer just a question of having a good product sheet or a good price: the algorithm needs to be convinced.
AIs now integrate external factors such as user-generated content (UGC), customer reviews and press mentions. The more active the brand’s ecosystem, the greater its legitimacy in the eyes of the agent. What’s at stake here is the ability to be readable and credible to a machine.
Finally, technical compatibility becomes crucial. Shopify already offers a native connector with ChatGPT, facilitating catalog integration. For other CMS, standardized APIs will soon enable direct connection of product feeds to AIs.
Delivery, a key experience lever
In the age of agentic commerce, delivery is becoming a strategic point of contact. In a purchasing journey now initiated elsewhere (AI, chatbot, conversational engine…), the moment the parcel is received often represents the first real contact with the brand. It is here that an essential part of the customer experience, and therefore of performance, is played out.
Logistics: the new AI selection criterion
Generative AIs don’t just recommend relevant products. They also integrate logistical criteria into their choice of merchants. Available stock, fast delivery, carrier reliability, post-purchase reputation and competitive price: all these factors influence the final decision.
A late parcel, a missing item or a broken promise are enough to generate negative reviews, and thus downgrade the brand in future AI recommendations. Today, over 60% of negative reviews concern delivery problems.
E-tailers therefore need to think ahead: making their upstream supply chain more reliable has become a prerequisite for gaining visibility.
Delivery as a brand experience
Well executed, delivery can, on the contrary, become a lever for differentiation. It’s an opportunity to materialize a promise, reinforce trust and create satisfaction. The Petrossian example is revealing: 24/48-hour delivery, cold chain maintenance, hand delivery, choice of time slot… The entire logistics system contributes to the brand’s upmarket image, and feeds its legitimacy in AI systems.
More generally, efficient delivery rests on solid foundations:
- Connecting sales channels and synchronizing inventory
- Automated order picking
- More delivery options: standard, express, relay
- Same-day shipping on eligible orders
- Real-time customer notifications
- Smooth returns management
These steps enable us to move from simple shipping to a true post-purchase experience, controlled from end to end.
Shippingbo, a logistical asset for emerging in AI
Shippingbo responds precisely to the requirements of this new context. Its software suite enables real-time stock synchronization, order routing to the right warehouses, and guaranteed D+1 dispatch, even during peak periods.
The results speak for themselves: +120% of orders shipped without hiring, 0 stock errors, guaranteed D+1 delivery. All signals that reassure AIs… and seduce consumers.
In an environment where choices are often made on the basis of a single answer, logistics are becoming an eligibility factor. More than ever, it conditions your presence… or absence in the results generated by artificial intelligence.
Capitalize on the post-purchase experience
With the rise of agentic commerce, re-purchasing is becoming a central strategic lever. While initial purchases can be generated by AI agents, loyalty remains a high value-added area for brands. And this requires a complete reinvention of the post-purchase experience.
Towards a new area of influence: post-delivery
Historically, e-commerce sites were primarily used for brand acquisition and discovery. Tomorrow, with single-product IA purchases, reassurance will shift to post-purchase: delivery experience, product use, customer service and ongoing relationship.
It’s a shift in the center of gravity: where pre-sales used to concentrate efforts, it’s now the post-sales phase that determines retention. Especially since, unlike marketplaces, LLM places the order… but leaves the customer relationship to you. A key difference that opens up a whole new world of possibilities.
An omnichannel, conversational strategy
In this new model, the e-commerce site becomes the crossroads of loyalty. It’s no longer just about selling, but about delivering a complete experience in several dimensions:
- Integrated conversational interfaces (intelligent search engines, advice, onboarding),
- Social and interactive content (videos, UGC, social proof),
- A synchronized CRM to maintain one-to-one relationships (WhatsApp, segmentation, retargeting),
- Enriched, experiential loyalty programs connected to omnichannel.
Tip: start structuring your data now and make it accessible (via API) to LLMs. The more readable your offer, the more you’ll be recommended.
Inspiration: when experience becomes a competitive advantage
Noise in the kitchen offers much more than a catalog: they cultivate a community of enthusiasts with content, recipes and online or in-store experiences. Their model is based on a strong trio: products + inspiration + customer relations.
QOQA goes further in product storytelling than the brands themselves, thanks to technological mastery, a singular tone and an active community.
Leroy Merlin makes DIY accessible to all, with a fluid phygital journey, AI advisors, and a promise of stocks per store.
Petrossian embodies service excellence, with vocal, embodied product information and flawless logistics.
Hugo Boss revolutionizes loyalty with a gamified token system, designed as a brand experience.
What was played out yesterday on your site is now being played out behind the scenes: logistics performance, data quality, post-purchase reputation… At a time when AIs are becoming the new showcases of e-commerce, brands no longer have a choice: they must perform across the entire chain, from promise to delivery.
To achieve this, companies need to rely on robust, integrated solutions: a OMS to orchestrate orders, a WMS to optimize inventory and picking, and a TMS to streamline shipments and deliver on customer promises. These technological building blocks are no longer incidental: they have become strategic to move up the list of AI recommendations.
This webinar brings together these two complementary visions to decipher the changes underway and suggest concrete courses of action. Whether you’re an e-merchant, marketing or logistics manager, you’ll leave with clear keys to anticipating, adapting and staying one step ahead.
👉 Couldn’t attend the live session?
Watch the replay now and turn the corner on AI-driven e-commerce with the right reflexes :
FAQ
This is a business model in which purchasing decisions are made by intelligent agents (AI) on behalf of the consumer. They search, compare, select and buy for them, according to their preferences and constraints.
AIs now evaluate logistical criteria (lead times, reliability, product availability) when recommending merchants. Fast, controlled delivery therefore indirectly improves your visibility in algorithmic suggestions.
These are software systems used to manage orders (OMS), stocks/warehouses (WMS) and deliveries (TMS) respectively. Together, they optimize the supply chain.
In a world where AIs buy, the first human contact with the brand becomes delivery. The post-purchase experience is therefore central to building loyalty and reassurance.
Because consumers (or AI) increasingly interact via natural interfaces: chatbot, voice search, messaging, etc. The site must therefore adapt to this fluid, dialogue-oriented navigation.
By working on your logistics, your content, your customer relations and your reputation. Brands need to be irreproachable on all fronts to meet the new AI criteria.
Glossary of terms to remember
Agentic commerce
An emerging business model in which an artificial intelligence acts in the customer’s place to research, select and purchase a product. The human expresses an intention, the AI executes the purchase.
OMS (Order Management System)
Order management software. It centralizes flows, synchronizes inventories in real time across all channels, and orchestrates the routing of orders to the right warehouses.
WMS (Warehouse Management System)
Warehouse management tool. It optimizes order preparation, location management, stock movements and logistics productivity.
TMS (Transport Management System)
Software solution dedicated to delivery management. It helps to select the right carrier, automate labeling, track packages and simplify returns management.
SEO logistics
Concept according to which logistics-related criteria (delivery times, available stock, carrier reliability) influence the visibility of merchants in AI recommendations.
Phygital
A contraction of “physical” and “digital”, this term refers to the seamless integration of physical (stores) and digital (website, app) channels in the customer journey.
Experiential loyalty
An approach to loyalty that goes beyond a simple points program. It relies on emotion, storytelling, exclusivity and quality of service to create a lasting bond with the customer.
LLM (Large Language Model)
AI-based language model, such as ChatGPT or Gemini. It is capable of understanding complex queries and generating natural responses. It serves as the basis for conversational agents.
Social commerce
The practice of selling directly via social networks or integrating social mechanisms into your e-commerce site (e.g. reviews, UGC, shares, etc.).
CRM (Customer Relationship Management)
All the tools used to manage and personalize customer relations. In particular, it enables you to centralize customer data, automate campaigns and optimize customer loyalty.
Package tracking
Real-time tracking of the status of an order, from preparation to final delivery, often communicated to the customer by email or SMS.
Logistics cut-offs
Daily cut-off times to guarantee same-day shipment. A key element in maintaining the delivery promise.
VIPization
The strategy of offering personalized, premium treatment to the best customers, to strengthen loyalty and satisfaction.