Model context protocol (MCP) provide rich context for event planning by defining which tools your AI platforms can access and how they should function. Rather than inputting data to your AI, MCP profiles pull data directly from your event management platform, saving your team time.
In an earlier post in our MCP profile series, we defined what MCP profiles are, outlined potential use cases, and explained how to start using them. In this post, you’ll learn how to prompt AI to make the most of your MCP profiles.
Make Prompts Specific
First, specificity is important. Your first prompt should clarify what platform you would like the LLM to draw information from and offer enough details — including dates, times, locations, and other attributes — to conduct the task efficiently.
Example:
Non-specific prompt:
“Help me build my event schedule, ensuring no speakers are double-booked.”
Specific prompt with details:
“Use RainFocus to help me organize my event schedule. Ensure no speakers are double-booked. Book the panels before noon each day. Don’t overlap sessions with similar topics. Schedule keynote sessions on the mainstage at 9:00 a.m., 11:00 a.m., and 4:00 p.m. for an hour each.”
Address Missing Event Data
Even the best written prompts will lead to a dead end if data is missing. For example, data on attendees, speakers, and exhibitor information is often partially entered. With the right prompts, MCP profiles can help you fill in gaps with customer and partner information from your CRM or CDP. Include instructions for how to treat empty form fields so that missing information does not create challenges with communication and reporting.
Example:
Data gaps not addressed:
“Pull a CSV list of all registered attendees for me to share with my marketing team.”
Data gaps addressed:
“Create two CSVs for me: one with the names and emails of all those who have registered and another with the names and emails of those who are partially registered.”
Define Output Format and Limitations
Though MCPs provide direct access to your data, your AI tool may not always structure responses exactly how you want them. Defining the length and format of your desired output improves the response. This is particularly important for use cases such as composing an email, filling in missing session information, or summarizing session feedback. Many LLMs default to paragraph format, which sometimes conflicts with character limits or overall readability.
Example:
Without the format defined:
“Provide me with a summary of attendees’ survey responses.”
With the format defined:
“Provide me with a bulleted summary of attendees’ survey responses. Each bullet point should be no longer than 350 characters. The summary should not exceed ten bullets.”
Learning to use MCP profiles effectively increases event teams’ productivity. Rather than exporting a spreadsheet or copying a codebase for various tasks, they can use RainFocus’ MCP profiles to publish AI-generated content (e.g., session details, abstracts, exhibitor information, registration forms, surveys) to the platform without leaving their LLM. Over the course of the event planning process, this capability saves significant time and effort.
