Agentic commerce

What is agentic commerce

To understand where we are headed, we need to distinguish between the AI tools we use today and AI systems that can take action on people’s behalf. Generative AI tools such as ChatGPT and Midjourney can create text, code, and images based on user input. Some AI tools can also use external tools and perform multi-step tasks, going beyond content generation.

In this scenario, agentic commerce is an emerging approach to buying and selling in which AI agents can research options, compare offers, and potentially complete transactions on a user’s behalf. Instead of manually handling every step, users can delegate tasks to AI agents within defined preferences, permissions, and limits.

AI agents

At the heart of this shift are AI agents. These are software systems designed to pursue specific goals by planning tasks, using available tools, and taking actions with varying levels of human supervision.

In the context of commerce, an AI agent can act as a personalized digital shopping assistant. A person can give it a budget and a goal (“find me a direct flight to Athens next Friday under 400$”), and it can search for options, compare prices, and present suitable choices. Depending on its capabilities and permissions, it may also complete the booking.

AI agents buying ads

AI ad buying agents may change how advertising campaigns are planned, managed, and optimized. Agentic advertising, rather than simply automating individual tasks, can potentially coordinate multiple steps, from analyzing campaign performance and identifying opportunities to supporting campaign setup and execution.

For publishers, the key question is how these developments could affect the way advertising inventory is discovered, evaluated, and purchased. AI agents may eventually play a greater role in matching advertisers with suitable inventory and coordinating transactions. However, these developments should not be confused with existing programmatic advertising, which already automates bidding and ad delivery. The extent to which AI agents will change the advertising market remains uncertain.

Agentic marketing

As AI agents become more involved in how people discover information, products, and services, publishers and marketers may need to adapt their strategies. Agentic marketing involves using AI agents to support or automate marketing tasks. For publishers, a related consideration is making their content accessible, understandable, and useful to AI-powered search and discovery systems. Here is your publisher homework to stay ahead of the curve:

• Double down on original, human-centric content:
  AI systems can summarize existing information, so publishers should focus on what makes their content distinctive: original research, first-hand experience, expert analysis, and trustworthy reporting;
• Build authority in your niche:
  Publish accurate, well-researched content that demonstrates expertise and provides genuine value to your audience. While no strategy can guarantee that an AI system will select or cite your website, distinctive and reliable content gives readers and information systems more useful material to work with;
• Optimize for AI-powered search:
  Traditional SEO remains important, but publishers should also consider how their content can be understood and surfaced in AI-generated search experiences. Use descriptive headings, clear explanations, relevant internal links, and appropriate structured data. Keep content accurate and well organized, and avoid assuming that FAQs or structured data alone will guarantee visibility in AI responses;
• Keep optimizing your ad monetization:
  As AI changes how people discover and consume online content, publishers should continue to monitor traffic patterns, ad performance, and revenue. Automated ad management and data-driven optimization can help publishers make better use of their available inventory. However, agentic commerce and programmatic advertising are distinct developments, and the impact of AI agents on digital advertising will depend on how these technologies evolve.