[MCP Basics: What It Is and What It's For] MCP stands for Model Context Protocol. It's a standardized protocol that connects an AI client to your Metricool account, letting the AI read and act on your scheduled posts and analytics in natural language instead of code. ## What is an MCP? An MCP is the bridge between an AI client and the data and tools it needs. It defines what data the AI can access and what actions it can take, so the AI understands your context without you repeating it. Think of it as an instruction manual for the AI that tells it: - What info it has access to - What goals it should focus on - What rules to follow - What tone to use (like casual or professional) - What to do if it doesn't know something ### Real-life example Say you want to connect your AI assistant (like Claude or another compatible client) to your Metricool account. The MCP tells the AI: > "You're helping someone who manages social media with Metricool. You can suggest content ideas based on real data, analyze post performance, and explain how scheduling works. If you don't have enough data, ask the user to connect their account." That way, your AI understands your context and gives you better answers without starting from scratch. ## What is it for? MCPs make AI more useful, accurate, and personalized. Instead of talking to a generic AI, you're chatting with one that gets your product, your role, and your goals. ### Example **Without MCP:** "How's my content doing?" — the generic AI replies, "Depends on the content. Can you give me more details?" **With MCP:** the AI already knows you're using Metricool, pulls your data, and replies: > "Your Instagram reels had 20% more engagement this week compared to last. Want some ideas to keep it going?" ### Key benefits - Smarter, more relevant answers - Fewer errors or confusion - Saves time — no need to repeat your context over and over ## What do you need to use Metricool's MCP? To connect your Metricool account to an AI using MCP, you need: - A **Metricool account** - A **client compatible with MCP** - A setup process — see [How to connect Metricool's MCP](https://help.metricool.com/how-to-connect-metricools-mcp-eqp9h), or the step-by-step guides for [ChatGPT](https://help.metricool.com/how-to-connect-metricools-mcp-with-chatgpt-0lrl3) and [Claude](https://help.metricool.com/how-to-connect-metricools-mcp-with-claude-0l84v) Metricool MCP works with any Metricool plan, including Free. Your plan limits still apply after you connect. > [!INFO] > **What's a "client"?** A client is an app that lets you chat with an AI model. It's where you type your questions; the MCP defines the context your AI uses to help you. ## What can Metricool's MCP do? Metricool's MCP connects your Metricool account with your favorite AI client. That means you can analyze data, create content, and automate tasks without leaving your workflow or copying and pasting information. ### Analytics and reporting - Check key metrics (followers, engagement, reach, and more) - Analyze posts by platform and format (Instagram, TikTok, Facebook, LinkedIn...) - Compare time periods (for example, February 2024 vs February 2025) - Generate reports with insights and tips ### Competitive analysis - See your defined competitors in Metricool - Compare growth, engagement, and posting frequency - Get strategy suggestions based on the data ### Content management - Schedule posts (even auto-translated to multiple languages) - Edit scheduled content - View your content calendar - Get best times to post ### Advanced automation with custom prompts - Extract content by campaign and date range - Spot patterns in high-performing content (formats, CTAs, topics) - Generate interactive reports - Compare formats (like Reels vs TikToks) - Create multilingual content across accounts with a single prompt ## Use cases - Compare your performance month over month - Analyze campaign impact - Find out what topics and formats work best - Post the same content across multiple languages and accounts - Discover why certain posts performed better than others ## Tools available in the MCP From your AI client (like Claude), go to the Integrations section to access these tools: - **Get available metrics** — Gather available metrics for a specific social network. - **Get Analytics Data** — Metrics from any connected social network, including competitor data. - **Get Best Time to post** — Best performing days and times for each network, based on your data. - **Get Schedule posts** — List all your scheduled posts, including already published ones. - **Create Scheduled Post** — Schedule posts across multiple brands and networks. - **Update Scheduled Post** — Update the text, date, or network of a scheduled post. > [!INFO] > The MCP currently covers your brands, analytics (including the analytics of your connected ad accounts), and the Planner. Inbox messages, post comments, Smartlinks, Google Business Profile reviews, autolists, Canva or Drive files, and the management of Ads campaigns (the Ads tab) are not available through the MCP at the moment. ## Quick summary If MCP is new to you, this table sums up the basics before you go deeper. | Concept | What it means | | --- | --- | | **MCP** | A protocol that defines the context an AI needs to help you properly | | **Used for…** | Making sure AI understands your tools, goals, and workflows | | **In Metricool** | We use MCP to connect your AI to your Metricool account with real context and data | ## MCP at a glance This side-by-side shows what changes when you use an MCP. | Feature | Without MCP | With MCP | | --- | --- | --- | | Prompt | Blind input | Prompt + structured context | | Memory | Very limited | Persistent and organized | | Data access | Manual | Direct and contextual | | Personalization | Minimal | Scalable and smart | ## Additional resources - 📹 MCP for Social Media: What It Is + What You Can Actually Do — [EN](https://youtu.be/ipyT3CJimgQ) · [ES](https://youtu.be/npfu_3d6FdA) - 📝 [What is MCP for Social Media Marketing?](https://metricool.com/what-is-mcp-for-social-media/) - _Check language availability using the article selector_ --- **Related articles** - [How to connect Metricool's MCP](https://help.metricool.com/how-to-connect-metricools-mcp-eqp9h) - [How to connect Metricool's MCP with ChatGPT](https://help.metricool.com/how-to-connect-metricools-mcp-with-chatgpt-0lrl3) - [How to connect Metricool's MCP with Claude](https://help.metricool.com/how-to-connect-metricools-mcp-with-claude-0l84v) - [FAQs about the Metricool MCP](https://help.metricool.com/faqs-about-the-metricool-mcp-1i3w0) - [Schedule and edit posts through an LLM with Metricool's MCP](https://help.metricool.com/schedule-and-edit-posts-through-an-llm-with-metricools-mcp-p8y4b) - [MCP vs API access: what is the difference?](https://help.metricool.com/mcp-vs-api-access-what-is-the-difference-5y3ib)