Skip to main content
You already use an AI assistant. Maybe it’s ChatGPT, Claude, Cursor, or something else. Now imagine it could see your actual Meta ad data, understand your business goals, and help you manage your campaigns. Not with generic advice, but with real answers based on your real numbers. That’s what AdAdvisor’s MCP instance does. It connects your AI assistant directly to your Meta (Facebook and Instagram) ad account so you can do things like:
  • “What’s my ROAS across all campaigns this week?”
  • “Pause every ad set spending more than $50/day with a ROAS below 2x”
  • “Create a new campaign targeting lookalike audiences with a $100 daily budget”
  • “Which ads are fatiguing? Show me the ones with rising CPMs”
Your AI reads your data, understands your goals, and takes action. You stay in control.

​
What does “MCP” mean?

MCP stands for Model Context Protocol. It’s an open standard that lets AI tools connect to external services. You don’t need to understand the protocol itself. Think of it like this: MCP is the plug, AdAdvisor is the adapter, and your Meta ad account is what’s on the other end. Your AI assistant uses this connection to read your ad data, run analyses, and (with your approval) make changes to your campaigns. No API keys to manage, no code to write, no developer setup.

​
What can your AI do once connected?

Read your data

Pull campaign, ad set, and ad performance. Ask about ROAS, CPC, CPM, conversions, and any other metric across any date range.

Understand your business

AdAdvisor knows your break-even ROAS, AOV, target CPL, and budget. Your AI doesn’t just report numbers. It knows what “good” looks like for you.

Manage your campaigns

Create, update, pause, and organize campaigns, ad sets, and ads. Adjust budgets, change targeting, swap creative. All through conversation.

Get recommendations

Ask for optimization suggestions grounded in your actual performance data and business context. Not generic tips. Real, actionable moves.

​
Works with the tools you already use

You don’t need to switch to a new app. AdAdvisor connects to whatever AI tool you’re already using:
  • Claude (Desktop and Code)
  • ChatGPT
  • Cursor
  • VS Code (GitHub Copilot)
  • Gemini CLI
  • Codex
  • Windsurf
  • Manus
  • Any other tool that supports MCP
Setup takes about 5 minutes. Most clients just need a URL and a quick login.

​
How is this different from using the Meta API?

If you’ve tried building your own Meta Ads integration, you know the pain. Developer apps, OAuth tokens, refresh flows, rate limits, API versioning, field selection. AdAdvisor handles all of that. You connect your Meta account once through a simple login flow. AdAdvisor takes care of authentication, token management, and data syncing. Your AI assistant gets clean, structured data designed specifically for LLMs, not raw API dumps. And because AdAdvisor stores your business context (margins, targets, goals), the AI can do more than query data. It can evaluate it.

​
Ready to get started?

1

Create an AdAdvisor account

Sign up at app.adadvisor.ai. You get a 7-day free trial with full access, including the MCP instance.
2

Connect your Meta ad account

Link your Meta account through a simple login flow. AdAdvisor syncs your campaign data automatically.
3

Connect your AI tool

Add the AdAdvisor MCP server to your AI client. Pick your tool and follow the step-by-step instructions.
4

Start asking questions

That’s it. Ask your AI about your campaigns and watch it work with your real data.

Full Setup Guide

Detailed walkthrough from account creation to your first query.

Connect Your AI Tool

Client-specific instructions for Claude, ChatGPT, Cursor, VS Code, and more.
Last modified on March 30, 2026