The AdRoll MCP server allows AI assistants such as ChatGPT, Claude, and other compatible systems to securely connect to your AdRoll account.
Once connected, you can ask questions about campaign performance, attribution, audiences, or account activity and receive answers based on your live AdRoll data. Instead of exporting reports or navigating multiple dashboards, marketers can explore insights directly through natural conversation.
This capability is built using Model Context Protocol (MCP), an open standard that enables AI assistants to connect securely to external platforms and systems.
What is Model Context Protocol (MCP)?
Model Context Protocol (MCP) is an open standard that allows AI assistants to securely connect to external platforms, APIs, and business systems.
Rather than copying data into a chat interface, MCP allows the AI system to retrieve structured information directly from connected platforms when it needs additional context. This allows AI assistants to move beyond static responses and interact with live data.
In practice, MCP acts as a bridge between large language models and real systems, enabling AI assistants to access information and workflows in a controlled, secure way.
What is the AdRoll MCP Server?
The AdRoll MCP server is a secure gateway that connects compatible AI assistants to the AdRoll platform.
When a user asks a question about campaigns or performance, the AI assistant can request the relevant information through the MCP server. The server retrieves data from AdRoll and returns it to the assistant, allowing it to generate a response based on the advertiser’s actual account data.
This allows marketers and teams to analyze performance, explore insights, and review campaign activity within the AI environments they already use.
The MCP server is different from the AdRoll AI Assistant, which operates directly within the AdRoll platform. The MCP server enables external AI assistants, such as ChatGPT or Claude, to access AdRoll data from outside the AdRoll interface.
If you want to learn more about using AI features within AdRoll itself, see Getting Started with the AdRoll AI Assistant.
What You Can Do
The AdRoll MCP server supports several workflows across the AdRoll platform.
Performance and Reporting Analysis
Retrieve campaign performance, analyze trends, and compare time periods.
Example prompts:
- “Show performance for the last 30 days.”
- “Which campaigns had the biggest week-over-week change in conversions?”
- “Summarize funnel performance trends.”
Optimization and Audience Insights
Identify underperforming campaigns and explore audience opportunities.
Example prompts:
- “Which campaigns are underperforming relative to budget?”
- “Identify audiences driving the strongest return.”
- “Where should I reallocate spend for better efficiency?”
Account Exploration (ABM)
For ABM programs, teams can explore account activity and intent signals.
Example prompts:
- “Which accounts have shown the highest intent activity recently?”
- “Are there any high-intent accounts that we haven’t targeted with a campaign yet?”
Sales Development Support (ABM)
Sales development representatives can identify accounts showing engagement or intent signals. This can help prioritize outreach and understand which accounts may be ready for a sales conversation.
Example prompt:
- “Tell me which accounts I should prioritize my outreach efforts for and help me draft an email.”
How the MCP Connection Works
The MCP connection includes two main components.
AI Client
The AI client is the application that runs the AI assistant and manages connections to external systems such as MCP servers. Examples include ChatGPT, Claude Desktop, or Cursor.
The client sends requests to the MCP server when the AI assistant needs external data and returns the results to the model, enabling it to generate a response.
MCP Server
The AdRoll MCP server acts as the secure bridge between the AI assistant and the AdRoll platform. It translates requests from the AI assistant into structured queries, retrieves the relevant information from the advertiser’s account, and returns it to the AI system.
When a user asks a question about campaign performance, the AI assistant recognizes that external data is required. It sends a request to the MCP server, which retrieves the relevant data and returns it to the assistant so it can generate a response.
Getting Started
The AdRoll MCP server is currently available through a closed beta program. If you are interested in testing this capability, please get in touch with our support team at support@adroll.com. Our team can help determine eligibility and provide configuration details and endpoints.
What You’ll Need
To participate in the beta, you will typically need:
- An active AdRoll account
- Access to an AI client that supports MCP connections
- Appropriate user permissions within AdRoll
Once approved for the beta, the AdRoll team will provide instructions for connecting your AI client to the MCP server.
Security and Permissions
Access to AdRoll data through the MCP server follows the same permissions model as your AdRoll account.
- The AI assistant can only access data available to your authenticated user account.
- All requests are scoped according to existing AdRoll permissions.
- Data is retrieved only when needed to answer a query.
During the beta phase, the MCP server is primarily optimized for data retrieval and analysis. Actioning capabilities will be added over time.
Troubleshooting
Authentication failed
Make sure you are using the correct AdRoll credentials and that your user role has Read or Admin permissions for the account you are connecting.
Missing data
If the AI assistant cannot find a specific campaign or account information, verify that you signed in to the correct AdRoll account during the authorization process.
You can also ask the AI assistant which AdRoll account it is currently connected to.
Feedback
The AdRoll MCP server is currently in a closed beta phase. The service is evolving quickly, and capabilities will change. Please provide feedback through this form.
Frequently Asked Questions
Does the AI assistant use my AdRoll data to train its models?
Data usage depends on the AI provider being used (for example OpenAI or Anthropic). We recommend reviewing the privacy and enterprise data policies for the provider used by your organization.
What data can the MCP server access?
The MCP server can retrieve campaign performance data, reporting insights, attribution information, and other supported account data available to your authenticated user account.
Is there a limit to how many queries I can run?
The MCP server follows AdRoll’s standard API rate limits. High-volume usage may result in temporary throttling.
Can AI assistants change my campaigns?
During the beta phase, the MCP server is optimized to retrieve and analyze data. Features for campaign creation or editing will be introduced in future updates.
Is this the same as the AdRoll AI Assistant?
No. The AdRoll AI Assistant operates directly inside the AdRoll platform.
The AdRoll MCP server allows external AI assistants such as ChatGPT or Claude to connect to AdRoll and retrieve account data.