# Meta’s ads MCP server comes to Databricks: Put your customer intelligence to work in advertising campaigns > Source: > Published: 2026-10-06 12:45:00+00:00 The ads MCP server from Meta is now available in Databricks Marketplace, connecting enterprise data and models to agentic advertising workflows. by [Katy Yuan](https://www.databricks.com/blog/author/katy-yuan), [Jen McNamee](https://www.databricks.com/blog/author/jen-mcnamee) and [Rohith Bhatia](https://www.databricks.com/blog/author/rohith-bhatia) Here’s a familiar scenario for enterprise advertisers: the data team has done real work. There’s a churn model in production, an LTV score that identifies customers worth investing in, and attribution data showing which creative drove revenue. Meanwhile, the media team is adjusting campaign budgets using the numbers in Meta Ads Manager. The insights that could change those decisions are still a handoff away: a CSV export, a custom Marketing API integration someone has to maintain, or a recurring Thursday sync where everyone hopes the findings make it into the next campaign. The [ads MCP server from Meta, now available in Databricks Marketplace](https://marketplace.databricks.com/details/311b6645-874c-4b67-9455-292d47fa7047/Meta_Ads-MCP-server-from-Meta), gives marketers a more direct path from those insights to campaign decisions. By connecting the ads MCP server to Genie, marketers can use natural language to explore Meta campaign performance alongside their governed business data in Databricks. The churn score, the margin table, and the campaign results can now inform the same conversation. This release expands our collaboration with Meta, following the launch of [Meta Conversions API on Databricks Marketplace](https://www.databricks.com/blog/activate-first-party-data-meta-conversions-api-databricks), which helps advertisers send first-party conversion signals from Databricks to Meta to improve ad measurement and optimization. MCP, or Model Context Protocol, is an open standard that lets AI agents interact with tools and data sources. The ads MCP server from Meta exposes 25+ tools across the campaign lifecycle: creating campaigns and ad sets, changing budgets and bids, defining audiences, managing creative and catalogs, pulling performance data, and diagnosing signal quality. Marketers already work in [Genie One](https://www.databricks.com/product/genie/one), where they ask questions and get instant insights grounded in enterprise context (including metrics and definitions curated by data teams), and take action across external tools. Adding Meta’s ads MCP server to Genie brings campaign reporting and advertising tools into the same conversation. Marketing and advertising teams can activate a model instead of exporting a segment. In Genie One, marketers can prompt: "Use our churn-risk score to identify high-value customers at risk of leaving, and launch a retention campaign with an $8,000 daily budget." Genie One then searches for the appropriate data, finds relevant enterprise context across connected tools and systems, and creates the campaign using a full understanding of how your business and marketing actually work. Or consider budget allocation; gross margin per order lives in Databricks, while campaign spend and delivery metrics live at Meta. An agent with access to both can recommend how to move budget toward ad sets driving profitable revenue, not just cheap conversions. If you’re already sending conversions to Meta through the Conversions API, the ads MCP server also gives agents access to signal diagnostics, including event volumes, Event Match Quality, and data freshness. “Did purchase events drop because sales slowed, or because something changed in our pipeline?” becomes an investigation that draws on both diagnostics from Meta and your data in Databricks. And the Monday morning question, “Which campaigns underperformed last week, and how does that square with our internal revenue?” becomes one prompt instead of stitching two tools together in a slide. What stops an agent from doing something expensive and inaccurate in a production ad account? In Databricks, administrators control access to the Meta connection through Unity Catalog, while [Unity Gateway](https://www.databricks.com/product/artificial-intelligence/unity-gateway) governs tool calls and records usage and audit logs. Enterprise data remains subject to existing Databricks permissions, and actions in Ads Manager. At Meta, advertisers set rules in Business Settings — block any budget increase over 20%, or disallow campaign creation on a given account entirely. The server checks every tool call against those rules before it executes. Break one and the call is blocked, with a structured error back to the agent saying why. This isn't a prompt politely asking a model to behave; it's enforced server-side, so it holds no matter which model you're pointing at it. And you can manage the same rules through the Marketing API, which matters if you're running hundreds of ad accounts and don't want to click through settings pages one by one. An LTV score, a propensity model, a margin table, and an inventory forecast each capture something useful about the business. Bringing that context into the same agent workflow as Meta’s advertising tools gives teams a way to use it in everyday campaign operations. Most integrations let an agent manage ads. In Databricks it can manage ads while reasoning over your whole data and AI estate — segments, propensity scores, LTV, first-party attribution, inventory. That's the difference between automating a few clicks and actually making a better call than the workflow it replaced. “Customers are asking us for a more direct way to put their data and AI investments to work in the decisions that drive growth. For advertisers, that means putting customer intelligence at the center of decisions about audiences, offers, and budgets. Our partnership with Meta helps customers connect the data and models they’ve built in Databricks to campaign execution, with the governance and control they need to scale.”—Stephen Orban, SVP Product Partnerships and Ecosystem at Databricks That same focus on putting customer intelligence to work extends to [CustomerLake](https://www.databricks.com/product/customerlake-cdp), an Agentic CDP built natively in Databricks that brings customer profiles, audiences, and campaign workflows into the enterprise data foundation. To get started, find the [ads MCP server from Meta](https://marketplace.databricks.com/details/311b6645-874c-4b67-9455-292d47fa7047/Meta_Ads-MCP-server-from-Meta) in the Databricks Marketplace and follow the instructions to configure a connection using a Meta user access token and the required permissions. Give your agent access to the relevant marketing data in Databricks, and begin with a concrete question your team already asks every week. Your teams have already invested in understanding their customers. This gives them a more direct way to put that understanding to work in Meta campaigns. If you have models in Databricks and a media team working off platform reporting, this is the shortest path between "we have the data" and "the campaign reflects it." [Get started with the ads MCP server from Meta in Databricks Marketplace](https://marketplace.databricks.com/details/311b6645-874c-4b67-9455-292d47fa7047/Meta_Ads-MCP-server-from-Meta). Subscribe to our blog and get the latest posts delivered to your inbox.