Foreword
Amazon has launched the Amazon Ads MCP Server in open beta in the US, and while it is not a visible change inside the Ads console yet, it represents a fundamental shift in how Amazon PPC can be managed at scale.
Paula Despins, Vice President of Ads Measurement at Amazon Ads, announced the launch during her keynote at IAB ALM on Monday, February 2, 2026. Rather than changing advertising strategy or fundamentals, this update expands what is possible operationally.
As of February 2026, this remains in open beta in the US, with no announcement yet on when it will become widely available.
The MCP Server is based on the Model Context Protocol (MCP), an open standard that allows AI agents to interact with Amazon Ads APIs in a more coordinated way.
As Despins explained in her keynote: "APIs remain essential for programmatic operations, but they were designed to expose specific capabilities one at a time, not to coordinate the kind of complete, autonomous workflows that AI agents now need to manage."
Rather than requiring AI agents to piece together individual API calls, the MCP Server acts as an instruction manual, guiding them through complete workflows. It also maintains automatic compatibility as Amazon's APIs evolve, eliminating the need for constant code updates or manual integration work.
The MCP Server allows AI tools like ChatGPT, Claude or Gemini to interact directly with Amazon Ads using natural language.
After setup, these AI agents can access Amazon Ads features including managing campaigns, checking performance reports, adjusting account settings, and viewing billing or financial data. Complex actions can be executed in a single flow instead of being split across multiple tools, dashboards or scripts.
In practice, this means tasks that would normally take multiple steps and touchpoints can now be triggered with a single prompt. For example:
Building full Sponsored Products campaigns from a simple brief
Instead of manually creating campaigns, ad groups, keywords and targeting, you can provide a brief such as: “Create a Sponsored Products campaign for this ASIN with a £50 daily budget.” The AI agent will set up the campaign structure, ad groups, and recommended targeting automatically, all in one step.
Generating custom performance reports on demand
Instead of exporting data, cleaning it and running analysis across multiple dashboards, a prompt like “Show me October’s performance for all campaigns with CTR and ACOS breakdown by product” delivers a structured, ready-to-use report instantly.
Rolling proven campaign structures into new countries quickly
Expanding campaigns into new marketplaces can be time-consuming. With AI agents, a prompt such as “Replicate my US campaigns in Canada, keeping budgets proportional” creates the campaigns in the new region, including targeting and structure, without repetitive manual work.
Optimising multiple accounts or brands at scale
For agencies managing multiple clients, AI agents can execute standardised updates, pausing underperforming ads, adjusting bids or duplicating successful campaign structures across accounts, reducing the risk of human error and saving hours of manual effort.
The biggest benefit is speed and consistency, particularly for agencies managing multiple accounts, brands or regions. Technical processes are streamlined, reducing complex integrations while maintaining operational control.
It's important to be clear about the limits of this tool. The integration of AI agents through the MCP Server does not replace the need for human expertise. This connectivity alone does not guarantee reliable outcomes, especially in advertising where real-world workflows often span multiple systems and decisions.
AI agents will not:
These remain core responsibilities for agencies and in-house teams. As Despins noted in her accompanying blog post, the MCP Server and its tools are designed to "reduce complexity by acting as an instruction manual, turning complex advertising operations into simple actions an agent can execute."
The focus is on enabling teams to concentrate on what differentiates them: their strategies, experience and expertise, rather than operational execution.
This update does not change what good Amazon PPC looks like.
It does, however, make it easier for experienced teams to move faster, launch smarter and scale what already works. Tasks that previously involved dozens of manual steps across multiple platforms can now be coordinated through guided workflows.
Brands working with agencies that understand both Amazon advertising strategy and emerging tools will be best placed to take advantage of these capabilities as Amazon continues to develop them.
Amazon is quietly upgrading the infrastructure behind its advertising platform. This is part of a broader move toward agentic capabilities that Amazon began introducing at its unBoxed event last year, now extending beyond the ads console itself.
For agencies, this is not about doing less. It is about being able to do more, across more markets, with better control and consistency.
If you want to explore how this could be applied to your Amazon advertising setup, our Marketplace team is happy to talk.
