How Calendly is using Rally's MCP to find prior research, enforce standards, and enable everyone

Watch the full webinar here.

When Ray Mason-Murillo joined Calendly as its first Research Ops hire, the problem was not a lack of research. It was a lack of visibility. Read Calendly’s case study here to learn about how Rally gave the team a central record to manage thousands of participants, maintain governance, and enable a self-serve practice. 

The MCP is what Ray turned to next – and he came at it from a completely different angle than most.

Starting with retrieval, not creation

Ray's first priority was not building studies. It was making existing research findable. Calendly had a disorganized research repository, and Ray wanted to know whether the MCP could fix it.

He picked a research topic, used transcript search to surface prior studies and specific interviews that relate. Rally’s MCP searched Calendly’s existing interviews to flag the strong matches and highlight when something looks like new ground. It came back fast and accurate with almost no fine-tuning. Ray verified the results himself, and they held up. 

"Immediately it was able to [pull interviews], with very little fine tuning, return that stuff really quickly, and it was high quality. At that point I went and double checked it myself, and it checked out. I was super impressed right away. This doesn't seem like a V1 at all. This is way, way beyond an MVP." - Ray Mason-Murillo

That capability matters because it changes how the team approaches new research. Before spinning up a study, anyone at Calendly can now ask what has already been learned on a topic, get a quality answer in seconds, and make a more informed decision about what questions are still worth asking.

Absorbing the operational overhead

The MCP also quietly took over work Ray used to do by hand. Study titles, naming conventions, indexing standards – the cleanup that fell to him whenever a designer or PM set up their own study. Now the agent names and structures studies the way Ray taught it, and the cleanup goes away.

The result is that more people can run their own research without Ray having to follow behind them. The standards are enforced by the system, not by a person.

Where Ray is taking it next

Ray's next goal is altitude. He wants C-suite stakeholders to be able to open Claude, ask a question in plain language, and find relevant prior research without needing to know how Rally works or log into a tool they will never open regularly. Meeting people where they already are, rather than asking them to learn something new.

‍

Read the full case study here.