How to Use an MCP Connector: A Step-by-Step Hotel Guide

Connecting AI to your hotel systems used to mean custom integrations, long implementation projects, and heavy IT involvement. That’s changing. With MCP (Model Context Protocol) connectors, hotels can securely connect AI assistants like Claude and ChatGPT to the software they already use, pulling live data, automating routine tasks, and getting faster answers without building a separate integration for every application.

This guide walks through the practical setup process for both Claude and ChatGPT, what you need before you begin, and how hotel teams can use MCP day-to-day. Whether you’re a revenue manager, front desk leader, or general manager, you’ll leave knowing how to get started and avoid the most common setup mistakes.

• MCP connectors provide a standard way for AI assistants to connect with your hotel’s existing systems. Instead of creating custom integrations, one connector can work across multiple compatible applications.
• Setting up Claude and ChatGPT is straightforward, but each platform has a slightly different workflow. ChatGPT also requires enabling Developer Mode before MCP apps can be used.
• Preparation matters. Before you begin, confirm your AI platform, verify that your PMS, CRS, RMS, or other software supports MCP, and make sure you have the necessary permissions.
• The biggest benefits come after setup. Hotel teams can use MCP connectors to retrieve live operational data, generate reports, answer guest questions faster, and support better revenue decisions from a single AI interface.
• A few simple best practices can prevent common issues. Testing the connection first, limiting access permissions, and verifying software compatibility will help ensure a smooth rollout.

What Is an MCP Connector? (A Quick Refresher)

An MCP (Model Context Protocol) connector is a standard way for an AI assistant to securely connect with other software and pull real-time information. Simply put, it’s the bridge between an AI tool and the systems you already use.

Think of it like a universal power adapter. Instead of building a custom connection for every property management system, revenue management platform, CRM, or booking engine, an MCP connector gives AI tools one consistent way to plug into many different systems.

For hotels, that means your AI assistant can pull the right information from connected platforms without a separate integration for every application. Once the connection is in place, you can focus on using AI to answer questions, automate tasks, and support daily operations, not the technical details behind the scenes.

What You Need Before You Start

A few minutes of preparation can save a lot of troubleshooting later. Before connecting an AI assistant to your hotel systems, make sure you have the right tools, compatible software, and the necessary permissions.

1. First, decide which AI platform you’ll use. Popular MCP AI connector options include ChatGPT and Claude, and each has a slightly different setup process.
2. Next, confirm that the hotel systems you want to connect, such as your PMS, CRS, or RMS, support MCP or offer an MCP connector. If you’re unsure, check with your software provider or IT team before moving forward.
3. Finally, identify who manages system access at your property. Many integrations require administrator approval or API credentials, so knowing who controls permissions can prevent delays during setup.

[TABLE PLACEHOLDER: Pre-Setup Checklist]

How to Use the Claude MCP Connector

Setting up the Claude MCP connector is straightforward and only takes a few minutes. Before you begin, make sure you’re using a Pro, Max, Team, or Enterprise plan, as MCP connectors aren’t available on the free version.

Setting Up the Anthropic MCP Connector

1. Go to Settings > Connectors in Claude.
2. Add the connector by entering its web address, or select one from the available connector directory.
3. Sign in to the connected system and approve the requested permissions.
4. Enable the connector inside the conversation where you want Claude to use it. Once it’s turned on, Claude can access information from that connected system during your chat.

A Simple First Task to Try

Start with a simple, low-risk request to confirm everything is working correctly. For example, ask Claude:

“Show me today’s occupancy summary for the hotel.”

This lets you verify that Claude can retrieve live data from your connected systems before moving on to more advanced tasks, such as analyzing revenue trends, identifying booking gaps, or generating operational reports.

How to Use the ChatGPT MCP Connector

The setup process for ChatGPT is similar to Claude, but there are a few important differences. Notably, OpenAI now refers to these integrations as apps rather than connectors, although you’ll still see the older term used in many guides and tutorials.

Setting Up the ChatGPT MCP Connector

1. Go to Settings > Apps and Connectors.
2. Turn on Developer Mode under Advanced settings. This step is easy to overlook and isn’t required when setting up Claude.
3. Add the app or connector by entering its web address, or install one from the Apps directory.
4. Enable the app inside each new chat where you want ChatGPT to access your connected hotel systems.

Once enabled, ChatGPT can securely work with data from your connected applications, allowing you to retrieve information, answer operational questions, and automate routine tasks without switching between multiple systems.

Claude vs. ChatGPT: What’s Different at Setup

Both Claude and ChatGPT support the same underlying Model Context Protocol (MCP). The main differences are in how each platform presents the setup process, not in the technology itself. If you know where to find the settings and how each interface handles connectors, getting started is straightforward with either tool.

Claude vs. ChatGPT Setup Comparison

Everyday Ways Hotel Teams Use an MCP Connector

Once your MCP connector is set up, the real value comes from using it in everyday operations. Instead of switching between multiple systems or manually compiling information, hotel teams can ask the AI assistant for answers using natural language and receive responses based on connected, real-time data.

Revenue managers can quickly pull live rate comparisons, review occupancy trends, or identify pricing opportunities without jumping between the RMS, CRS, and market data sources.

Front desk staff can get instant answers about room availability, arrivals, departures, or reservation details, helping them respond to guest questions faster and reduce time spent navigating multiple applications.

Operations and leadership teams can generate daily performance reports, occupancy summaries, and other routine updates without manually exporting and combining data from different systems.

MCP Connector Use Cases by Role

Common Mistakes to Avoid

A few simple checks can prevent most setup and usability issues. Keep these common mistakes in mind:

• Giving the AI tool more access than it needs: Follow the principle of least privilege by granting only the permissions required for the tasks you want it to perform.
• Skipping a test run: Before relying on the connector for daily operations, try a simple request to confirm the connection is working and the data returned is accurate.
• Assuming every hotel system supports MCP: Not all PMS, CRS, RMS, or other applications are MCP-compatible. Verify compatibility with your software vendor before you begin.
• Forgetting to enable Developer Mode in ChatGPT: This is a required step for using MCP apps in ChatGPT and is easy to overlook during setup.

Where This Is Headed

As more hotel teams make “just handle it” a normal way of working with AI, connected technology stacks and revenue tools like Lybra Assistant RMS are increasingly designed to take advantage of secure MCP-enabled workflows that simplify operations and support better decisions.

With SimpleBooking’s MCP connector, our clients can already connect their booking technology to AI-powered workflows and put these capabilities into practice. By enabling secure, connected access across systems, the connector helps hotel teams reduce manual steps, surface relevant information faster, and let AI handle routine tasks while teams stay focused on delivering better guest experiences.

Want to learn how a connected PMS, CRS, and RMS can reduce service friction across your property? Explore the Zucchetti North America technology stack:

• Vertical Booking CRS
• Lodgical Solution PMS
• TCPOS
• Lybra RMS
• Simple Booking IBE

Have questions about where to start? Contact our team, and we’ll help you find the right fit.