AI agents like ChatGPT, Claude, and Gemini have promised a lot in the way of simplified workflows. We’re all for it. The issue is, there’s plenty of room for error in the way many of us feed them data. 

Run a report. Save the file. Upload it to your agent.

Now you have answers, assuming the data’s right. The agent will give you the best answer it can, even if it’s lacking essential data points. And that saved file is a copy of your data, sitting in the account you uploaded it to. What happens tomorrow? Do you upload a sales report every day? Every hour?

We built the Encore Pro MCP so your agent can ask Encore Pro directly, anytime you ask it a question. It works within each user’s permissions and scoping, without access to customer personal information. We keep a record of every action it takes.

‍

What is an MCP?

A Model Context Protocol (MCP) is a way to give AI agents access to tools and data sources beyond their default knowledge. 

Without a proper MCP, an AI agent will try to answer questions to the best of its ability, but it doesn’t have the full picture. By connecting it to a data source like Encore Pro, you can feel confident that your agent has the information that you do.

‍

Why not just upload a report?

Things move fast in the events world. Sales are changing and customers are engaging with campaigns. That’s one of the key ideas behind Encore Pro: we believed that live dashboards were, in many cases, far more useful than reports run at a fixed point in time.

At the risk of carrying a metaphor too far, those dashboards are like the MCP. Instead of running a report and verifying the criteria, scoping, context, security, and sensitivity, you can now connect your agent directly to the source of the data. 

The Encore Pro MCP follows all user-based scoping (so, if a user can only see certain shows, their MCP can’t either) and does not share Personal Identifying Information. Every call that the AI agent makes is logged, and, your organization can put limits on which agents (or even which accounts on which agents) can connect to your data.

As always, when using AI providers, there are some important caveats. Make sure you know the details of your data processing agreement with your AI provider, and how they’re retaining copies of this data (if at all).

‍

Why build this MCP?

Most of us now use AI to answer questions at work. We built the MCP to do this in a way that protects your data while creating fewer error points.

People are already doing this, and it’s easy to see why. You generate a report in your ticketing system, and you hand it to an agent because that’s quicker than staring at a spreadsheet. 

Nothing about that is careless. The trouble is that it’s  clumsy, and there’s no way to know afterward what the agent had in front of it.

If connecting your agent to Encore Pro is simpler than exporting a report, more people will use that secure connection, and fewer will try the workaround.

Encore Pro enforces permissions in one place, so we can give an AI a set of permissions and know that whatever it tries, it can’t go past them. We didn’t have to invent a new way to govern data, just use the governance we had already built for dashboards, reporting, segmentation, lists, and revenue. 

The same setup makes it traceable. Every request comes through the same door, so we record each one: which show, which dates, and whether it worked. That means what the agent saw can be worked out afterward, which a pasted report can never tell you.

‍

Want to get started?

Reach out to our Client Success team, and they’ll walk you through what setting this up will look like.