How to Set Up Supermetrics AI for Marketing Reports

What Supermetrics AI Does

Supermetrics has been a marketing data pipeline tool for over 15 years, pulling metrics from 150+ platforms into spreadsheets, dashboards, and warehouses. In 2025 and 2026, the company added a layer of AI on top of that pipeline. Instead of exporting data to a spreadsheet and manually building charts, you can now ask questions about your campaigns in plain English and get answers, charts, and recommendations back.

The AI layer works in two ways. First, built-in agents inside the Supermetrics interface analyze your connected data directly. Second, integrations with external AI assistants (Claude, ChatGPT, Microsoft Copilot, Google Gemini) let you query your marketing data from within tools you already use. There is also a Model Context Protocol (MCP) server that gives AI agents real-time, governed access to your marketing platforms without custom API code.

Prerequisites

Before you start, you need three things:

1. An active Supermetrics subscription. All packages from Starter to Enterprise now include Data API access, which powers the AI features.
2. At least one connected data source. Supermetrics supports Google Ads, Meta Ads, LinkedIn Ads, TikTok Ads, GA4, HubSpot, Salesforce, Shopify, and dozens more.
3. AI chats enabled as an additional destination in your plan. Check your Supermetrics Hub settings to confirm this is active.

If you already use Supermetrics for spreadsheet or warehouse reporting, you likely have the first two covered. The AI destination may need to be activated in your account settings.

Step 1: Connect Your Data Sources

Log in to the Supermetrics Hub at hub.supermetrics.com. If you have not connected your marketing platforms yet, do that first. Click "Add new data source" and authenticate with each platform you want to query. The most common starting set is Google Ads, Meta Ads, and GA4, which covers paid search, paid social, and web analytics in one shot.

Each connection requires OAuth authentication with the respective platform. You will be redirected to the platform's login page, where you grant Supermetrics read access to your data. No write permissions are needed.

Once connected, Supermetrics pulls in your account structure (campaigns, ad groups, ad sets) and begins syncing historical data. The initial sync typically takes a few minutes depending on account size.

Step 2: Enable the AI Destination

In the Supermetrics Hub, navigate to "Destinations" and look for the AI assistants section. This is where you activate the bridge between your marketing data and your preferred AI tool.

For Claude, click the Claude integration card and select "Try for free" or "Connect." You will be directed to authorize the connection between your Supermetrics account and your Claude workspace. The integration uses Supermetrics' own connector infrastructure, so your data flows through their pipeline with the same transformations and quality checks applied to your regular reports.

For ChatGPT, the process is similar. Visit the ChatGPT integration page from the Hub, click "Try for free," and follow the authorization flow. ChatGPT is recommended for ad-hoc queries and quick trend identification.

Claude is optimized for longer strategic analysis, multi-step audits, and detailed reporting. Choose based on your typical workflow.

Step 3: Ask Your First Question

Once connected, open your AI assistant and start with a simple query. Good starting questions include:

"What was my Google Ads spend last month compared to the month before?"

"Which Meta ad set had the highest cost per lead this week?"

"Show me a breakdown of GA4 sessions by source for the past 30 days."

The AI assistant will query your connected Supermetrics data in real time, apply any transformations you have configured, and return the answer with context. For example, rather than just saying "leads dropped 15%," the Insights Agent provides context like "leads dropped 15% because your top campaign in Germany paused on March 12."

Start simple and get more specific as you build confidence in what the integration can handle.

Step 4: Use the MCP Server for Advanced Workflows

For teams that want deeper integration, Supermetrics offers a Model Context Protocol server. MCP is an open standard that lets AI agents connect to external data sources through a standardized interface. The Supermetrics MCP server acts as a universal bridge between your AI tools (Claude, ChatGPT, Gemini, Copilot) and your marketing platforms.

The advantage of the MCP approach is that it gives AI agents governed, real-time data access without requiring you to write custom API code or manage authentication for each marketing platform individually. The MCP server handles connection management, rate limiting, and data normalization.

To enable it, check the Supermetrics documentation for MCP server setup instructions specific to your AI platform. The configuration typically involves adding the Supermetrics MCP endpoint to your AI tool's connection settings.

Step 5: Explore the Insights Agent

The Supermetrics Insights Agent is a built-in AI assistant that runs inside the Supermetrics platform itself. It examines your connected marketing data and surfaces patterns, anomalies, and recommendations automatically.

The agent supports natural language queries in multiple languages, including English, French, Spanish, and Japanese. It can generate dashboards on the fly, flag underperforming channels, and suggest optimization actions.

Currently, the Insights Agent works best with Facebook Ads, Google Ads, and Google Analytics data. Support for additional sources and cross-channel blended analysis is being expanded. No technical setup is needed beyond having your data sources connected. Just open the agent and start asking questions.

What Supermetrics AI Handles Well

The platform works best for three use cases. First, regular performance check-ins where you want quick answers about spend, leads, ROAS, or traffic without opening multiple dashboards. Second, anomaly detection where the AI flags something you might miss in a weekly report, like a sudden CPC spike or a campaign that quietly stopped delivering. Third, cross-platform comparisons where you want to see Google Ads and Meta Ads performance side by side in a single natural language query.

The AI layer is particularly useful for marketing managers who report to leadership. Instead of spending hours building slide decks from exported data, you can ask the AI assistant to summarize performance, identify the top three wins and losses, and generate a formatted overview.

Limitations to Know

Supermetrics AI is strong on marketing data but does not replace a general-purpose analytics tool. It works with the data sources Supermetrics connects to, which are primarily advertising, analytics, CRM, and ecommerce platforms. If your analysis requires blending marketing data with internal databases, financial systems, or custom data warehouses, you will need to handle that layer separately.

The Insights Agent's cross-channel blended analysis is still being developed. For now, queries work best when targeting a single data source at a time. Multi-source analysis is available through the AI assistant integrations (Claude, ChatGPT) but may require more specific prompting.

If your reporting needs extend beyond marketing into operational data, financial modeling, or multi-department analysis, a tool like VSLZ can handle file uploads and multi-source data analysis from a single prompt without requiring pre-configured connectors.

Practical Summary

Setting up Supermetrics AI takes under ten minutes if you already have connected data sources. Enable the AI destination, authorize your preferred assistant, and start querying. The MCP server adds a layer of flexibility for teams building automated workflows. For most marketing teams, the fastest path to value is connecting Google Ads, Meta Ads, and GA4, then asking the AI assistant to run through last week's performance. You will have a usable summary in seconds instead of hours.