In recent years, e-commerce has focused particularly on personalisation, recommendation algorithms, and customer service through chat. We are now entering a phase where the role of AI shifts from being an assistive tool to a more independent entity. This development trend is called Agentic Commerce.

What is Agentic Commerce?

Agentic Commerce refers to the transition from passive online services to proactive AI agents.

Until now, purchasing has required active work from the user: searching, comparing, reading product information, and completing the process. In Agentic Commerce, part of this work transitions to AI agents, which can make decisions and perform tasks independently within given frameworks.

AI agent is not just a conversational language model. It is software capable of understanding objectives, leveraging data, and executing transactions on behalf of the user or merchant.

How does Agentic Commerce change the buying experience?

AI agents streamline trade on three different levels:

1. Agent as a consumer aid

Shopping becomes easier when an AI agent can handle routine tasks. Instead of the customer themselves searching for suitable spare parts or comparing insurance conditions, they can assign the agent to find the best option according to budget and need. The agent prepares the shopping cart or completes the order within authorised limits.

2. Agent as a merchant's tool

For a merchant, agents mean more automated processes. An agent can, for example, optimise stock levels based on demand forecasts or manage dynamic pricing in real-time, saving specialists time for more strategic planning.

3. Trading between agents (B2B)

In the B2B sector, the potential is vast. In the future, the purchasing system of a buying company could communicate directly with the supplier's system, negotiate delivery terms, and confirm orders automatically (or involve a human at this stage) when certain criteria are met. This removes unnecessary intermediaries and reduces human errors. We are not there yet, but the direction is clear.

Why is this topic relevant right now?

Agentic Commerce is not a distant vision but a logical continuation of several key developments:

• The capability of language models: AI now understands even broad contexts and can perform logical reasoning.

• The prevalence of interfaces: Modern, API-based e-commerce platforms make it possible for AI to communicate with other systems seamlessly.

• Common standards: For agents to operate efficiently between different marketplaces and systems, they need a common protocol. At the heart of this is the Universal Commerce Protocol (UCP), which creates a framework for how trade-related information and transactions are transmitted between systems seamlessly. Read more about what Universal Commerce Protocol means in practice.

How to prepare for this development?

From Crasman's perspective, the spread of agents highlights the importance of a digital foundation. If you want your company to be ready to serve AI agents, consider these:

1. Structured product information: Agents need clear and machine-readable data. Invest in quality product information management (PIM and product information management).
   
   

2. API readiness and Headless model: Ensure that your e-commerce architecture allows external agents to interact with it easily (see Headless systems).
   
   

3. Building trust: Secure payment methods and clear rules on what agents can do are critical for customer satisfaction.

Agentic Commerce is a natural step towards a more effortless digital transaction. It does not replace the traditional interface or visual experience, but it removes unnecessary friction from the purchasing process and diversifies what e-commerce means.