Amsterdam, 9 July 2026 – AI agents that independently set up campaigns, configure targeting, and pull
reports are no longer a thing of the future. adpaq, the Dutch Demand Side Platform (DSP) for programmatic
digital audio, has developed its own Model Context Protocol (MCP) server. This allows AI assistants to
communicate directly with the platform and carry out campaign tasks using natural language.

Together with Abovomaxlead, adpaq is the first to bring this technology into practice within programmatic digital audio. Over the coming period, both parties will test how agentic AI can speed up and simplify the campaign process. After the validation phase, the functionality will become available to all of adpaq’s partners.

Where generative AI today mainly supports the writing of text or the analysis of data, agentic AI goes a step further. AI agents can independently perform multiple actions to achieve a goal. Within advertising, this means, for example, setting up a campaign, selecting targeting, linking creatives, and generating reports.

This development calls for a new way for AI to communicate securely with advertising platforms. The Model Context Protocol (MCP) is the emerging open standard for this purpose. Through this protocol, AI agents can not only retrieve information but also carry out controlled actions within external systems.

With its own MCP server, adpaq enables AI agents to work directly within the DSP. Based on a briefing, an agent will soon be able to prepare a complete audio campaign, create line items, set targeting by location, device, publisher, or audience, link creatives, and analyze performance. The user always retains control.

“For adpaq, this is a logical next step. We see every day how agencies and advertisers want to be able to move faster, and agentic workflows fit that perfectly,” says Marvin Vacquier Droop, Founder & Chief Product Officer at adpaq. “By building our own MCP server, we retain control over how AI agents interact with our platform, rather than waiting to see what others decide for us. That reflects how we at adpaq have always worked: leading the way rather than following.”

According to adpaq, agentic AI requires more than just a technical integration. That is why several layers of security have been built into the MCP server. Informational tasks can be carried out entirely by an AI agent, while sensitive actions, such as budget changes or deleting campaigns, always require explicit approval from the user. “Especially in advertising, you want to combine speed with control,” Marvin continues. “An AI agent should be able to speed up your work, but should never make unexpected decisions.”

To test the technology under realistic conditions, adpaq is working with Abovomaxlead, the largest independent media and marketing agency in the Netherlands. As the first implementation partner, Abovomaxlead is exploring how agentic AI can be deployed within existing campaign processes. Both parties are looking not only at speed, but also at reliability, ease of use, and how AI can best collaborate with campaign specialists. “AI only becomes truly valuable when it helps our people work faster and smarter. Together with adpaq, we are exploring how agentic AI can take over routine tasks, so our specialists have more time for strategy, creativity, and delivering impact for our clients,” says Sander van der Peet, Product Lead Audio at Abovomaxlead. Over the coming period, the technology will be tested extensively across various campaign and reporting scenarios. After the summer, adpaq will make the MCP server available to all affiliated agencies and advertisers.

Internationally, an increasing number of technology companies are investing in agentic AI as the next development in digital advertising. According to adpaq, this shifts the role of AI from a supporting assistant to an active executor of routine tasks. With the introduction of its own MCP server, adpaq responds to this development and makes agentic workflows available to the Dutch programmatic digital audio market — not to replace the expertise of marketers, but to give them more time for strategy, creativity, and achieving better campaign results.