Google brings a Gemini-powered agent to Ad Manager

If you have ever worked in ad operations, you know the specific pain of an underperforming campaign: opening six browser tabs, exporting clunky CSVs, and hunting for a broken targeting filter for two hours.

Google wants to hand that headache over to AI. The company announced Ask Ad Manager, a conversational agent built on Gemini specifically designed for ad publishers.

The tool enters beta testing this month, with more capabilities rolling out across the rest of the year.

Talk to your ad inventory

Ask Ad Manager basically turns Google's notoriously complex publisher dashboard into a conversational chat window.

First, it handles real-time troubleshooting. Instead of manually parsing delivery reports to see why a line item is underdelivering, publishers can ask the assistant directly, get step-by-step diagnostic advice, and ask follow-up questions to restore stalled revenue.

Second, it generates custom reports from natural language prompts. If you need a quick benchmark comparison or a specific view of inventory performance, you just ask, and it produces the table or report on the fly.

Third, it navigates the interface for you. Ad Manager has thousands of deeply nested settings. The agent will drop direct links into the chat that take you straight to the right page, automatically pre-loading the relevant filters and configurations based on your prompt.

Google was quick to emphasize that Ask Ad Manager runs exclusively on each publisher's own data, keeping proprietary ad metrics isolated.

Yahoo, MCP, and autonomous ad agents

The bigger story lies in where Google is heading next. This is not just a chat widget on a dashboard.

Publishers like Yahoo are already wiring Ad Manager into their own internal custom agents to automate forecasting, line item creation, and ongoing campaign optimization.

To support that kind of automation, Google revealed that it will launch dedicated REST APIs and an MCP server later this year. Adopting Anthropic's Model Context Protocol is a notable nod toward open interoperability for autonomous tools.

Google is also developing specialized agents for both publishers and media agencies to automate inventory discovery, negotiation, and execution, alongside infrastructure meant to let first-party and third-party agents transact at scale.

Why it matters

For publishers and digital ad teams, this tool should offer an immediate productivity boost. Google Ad Manager is an essential but notoriously frustrating piece of software. Cutting out the manual labor of report stitching and broken line item debugging is an obvious win, and publishers should jump into the beta as soon as it opens.

For the broader ad tech industry, the moves around MCP and agent-to-agent negotiations are the real game changer. We think Google is quietly laying the groundwork for the next era of programmatic advertising, where buying and selling ads shifts from human traders to collaborating autonomous agents.

The catch, however, comes down to execution risk. Ad trafficking is unforgiving: an AI hallucination on a budget cap, flight date, or audience exclusion can burn thousands of dollars in minutes. Giving an agent conversational controls is smart and safe; letting autonomous third-party agents pull the levers across the campaign lifecycle will require a level of trust that the ad industry does not yet have.