Dive Brief:
- Amazon is beta testing a new platform that integrates Amazon Ads with a range of artificial intelligence platforms to enable simpler, coordinated campaign integration, the company announced at IAB ALM last week.
- The new Amazon Ads MCP Server is built on the Model Context Protocol, an open standard enabling AI platforms to communicate with external tools. The Amazon Ads MCP Server acts as a translation layer between AI agents and the Amazon Ads API, turning natural language prompts into structured campaigns.
- Additionally, the Amazon Ads MCP Server includes pre-built capabilities and workflows to reduce the complexity of launching and running Amazon Ads campaigns. The tools reduce multistep workflows, like creating accounts, generating reports and expanding campaigns into new locales, into simple actions.
Dive Insight:
It seems like every new day brings another AI-powered platform geared toward making campaigns easier to create, launch and track. Ironically, with so many tools and agents in the market, managing them has become extraordinarily complex. The Amazon Ads MCP Server is designed to enable multiple AI platforms to seamlessly integrate in a common marketplace.
Specifically, the Amazon Ads MCP Server recognizes the limitations of current APIs, which were designed to handle discrete tasks individually rather than the complete workflows AI agents are responsible for managing. As a result, these agents are piecing together individual steps, rather than running a coordinated campaign, reducing efficiency and effectiveness.
The Amazon Ads MCP Server acts as an “instruction manual” for AI agents to turn the complex, multistep operations of advertising workflows into simple actions that can be handled through natural language prompts, per a blog post from Paula Despins, vice president of ads measurement at Amazon Ads. The result is a simplified, easier-to-use platform.
“For example, if an advertiser is running campaigns in the U.S. and Canada, we have a tool that enables them to quickly expand it to another country with a single prompt,” Despins wrote.
A marketer could also use the server’s built-in tools to create an end-to-end Sponsored Products campaign – a process that typically requires at least three operations, including creating the campaign, setting up ad groups and creating the ads – using a single prompt.
The server works with both custom-built agents and platforms such as Claude, ChatGPT and Gemini to help unify insights and automations. It is also designed to be compatible with Amazon’s APIs as they evolve, without requiring code rewrites or other maintenance.
The Amazon Ads MCP Server builds on an Ads Agent tool that Amazon introduced at its annual unBoxed event. That tool made it possible to use conversational language to help agentic AI work with Amazon’s first-party data within the Amazon Ads console.
Amazon saw revenue increase 14% to $213.4 billion in the fourth quarter of 2025, per its recent earnings report. AI and its Amazon Web Services business was a focus of the report, in which the company pledged to invest about $200 billion in capital expenditures, predominantly in AWS.
