A purchasing director who defines their needs on ChatGPT before even consulting a supplier’s website, software agents negotiating and placing orders without human intervention, self-service portals handling nearly half of transactions: the B2B landscape has changed dramatically in just a few months. Companies that do not adjust their customer journey and data strategy risk losing visibility at the very moment the buyer begins their search.
B2B Buying Journey and AI Assistants: The Starting Point Has Changed
When discussing B2B innovation, CRM or automation often come to mind. However, the most concrete change is happening upstream: buyers are consulting AI assistants before visiting a supplier’s site. ChatGPT, Gemini, and Perplexity are used to define needs, compare offers, and structure the purchasing process.
Recommended read : The latest trends and innovations to discover in the world of Swiss startups
For a company selling services or products online, the consequence is direct. If your content is not structured to be read and retrieved by a language model, you no longer exist in the discovery phase. We are talking about structured data, clear formats (comparison tables, tagged FAQs), and pages that answer a specific question rather than corporate brochures.
Following the news from Direct B2B allows you to spot these behavioral changes over time, before they become obvious.
Read also : How to remotely follow the LesEntreprisesFrenchy 2026 event and discover its full program
This shift towards an “AI-first” journey does not eliminate human contact. It shifts it. The buyer arrives better informed, with a pre-formed shortlist. The salesperson’s role starts later in the cycle, focusing on personalization and negotiation issues, not product education.
AI Agents in B2B Purchasing: Procurement Automation
The next step is more radical. AI agents are beginning to take over complete purchasing sequences: supplier identification, price comparison, order placement. According to an analysis dedicated to AI agents in B2B procurement, purchase cycles are shrinking from several weeks to sessions completed in just a few minutes.
Gartner anticipates that by 2028, 15 billion connected products will be able to autonomously purchase services and trigger orders. We are no longer talking about a distant future, but about an infrastructure that is already being deployed in the supply chain.
What This Changes for Suppliers
If a software agent can compare three offers in seconds, the readability of your online catalog becomes a competitive advantage. Feedback varies on the actual maturity of these agents across sectors, but a few technical points are consensus among companies preparing for this:
- Expose open APIs with standardized product data (price, lead times, availability) so that the agent can query your catalog without going through a graphical interface
- Keep product sheets updated with standard identifiers (GTIN, customs codes) to avoid being excluded by automatic sorting
- Provide machine-readable commercial terms, not buried in a 40-page PDF
The business model itself is evolving: when the buyer is no longer a human navigating, the notion of “customer journey” transforms into a data journey.
Self-Service Portals and B2B Customer Experience
In the B2B e-commerce space, the trend towards self-service is not new, but its scale is accelerating. According to an Akeneo study reported in the sector, 49% of B2B transactions now go through self-service portals. Orders can be placed, tracked, and replenished without picking up the phone.
For field teams, this means rethinking the architecture of the client portal. A good B2B portal in 2026 no longer resembles a showcase site with a contact form. It integrates account management, order history, client-specific pricing customization, and increasingly often, a recommendation engine powered by purchase data analysis.
Content Personalization and Engagement
Hyper-personalization, borrowed from B2C, is becoming an expected standard. Prospects want to see their negotiated terms, their usual references, and relevant suggestions as soon as they log in. A B2B portal that displays the same generic catalog to all its clients loses engagement.
Data is the fuel for this personalization. However, many companies collect data without leveraging it, due to a lack of tools or internal analysis skills. The priority investment is not always a new marketing tool; sometimes it is cleaning up the database and establishing actionable segments.
B2B Content Strategy Adapted to Engines and LLMs
Content production remains a pillar of B2B marketing, but the format is changing. Publishing a 30-page white paper and waiting for leads is no longer sufficient when the buyer can obtain a summary in three seconds via an AI assistant.
What is currently working in the field:
- Short, structured content by question-answer format that language models can extract and return with attribution
- Comparison tables with up-to-date data, more useful than a descriptive paragraph for queries like “which tool to choose for…”
- Quantified and dated case studies, serving both classic SEO and credibility with LLMs
- Industry events (webinars, conferences) whose content is republished in an indexable format, not just as video replays
Optimizing content for LLMs does not replace classic SEO, but adds to it. The two channels coexist, and companies that work on both capture a larger share of the buying journey.
B2B is going through a period where tools are changing faster than organizations. AI agents, self-service portals, and the redesign of content for language models are not isolated projects. They form a coherent whole: making the offer readable by machines as well as by humans. Companies that advance on these three fronts simultaneously gain a concrete, measurable advantage as soon as the next buying cycles.