How Ticketmaster Cut Time-to-Insight with Databricks Genie One

Ticketmaster’s mission is to connect millions of fans with the events they love, and every high-demand ticket sale depends on fast decisions. Operations and client teams need quick answers from live and historical data to help organizers deliver a smoother sale and a better fan experience, without waiting on SQL experts or submitting data requests. By building Wizard, an internal chat experience powered by Genie One on the Databricks Platform, Ticketmaster put governed, self-serve insights directly into the hands of the teams supporting these live ticket launches.

Data access bottlenecks stalled real-time onsale decisions

At Ticketmaster, a live onsale is a time-sensitive window where inventory, demand and show performance shift in real time. In these moments, access to the right information isn't just helpful - it directly impacts outcomes.

Ticketmaster had already done the hard work of capturing historical data. The team was saving snapshots of peak sales every 15 seconds across more than 60 metrics, building a rich historical record that operations and client teams could learn from. The problem was turning that data into answers quickly enough for people who needed them in real time.

There was a bottleneck accessing data for those who needed it most. The majority of business users lacked a direct connection to the underlying data, and those who did were often unfamiliar with SQL. Critical signals lived across Databricks, Splunk, and other systems, forcing teams to track down the right data owner and file a request. In proof-of-concept interviews, two to three source data teams each reported losing a full sprint per quarter, roughly two weeks, to ad hoc requests. The deeper cost was invisible. “Every time someone had to submit a request, it introduced delay,” said Brian Muehlenkamp, Sr. Product Manager for Demand Intelligence at Ticketmaster. “As a result, the most valuable questions never got asked at all.”

How Ticketmaster built Wizard with Databricks Genie One

Ticketmaster needed a way to put answers in the hands of operations and client teams without locking them behind a SQL query or a ticketed request. The team built a working prototype and pitched leadership within weeks, and the project moved fast enough to be pulled into a quarter it was not originally planned for. That proof-of-concept became Wizard, Ticketmaster's internal AI chat experience powered by Genie One. Early adoption followed a deliberate path: questions were funneled through the workspace internally first, then access was opened to key stakeholders, and once the value was clear, engineering stepped in to embed it in the product.

Instead of waiting on a dashboard update or escalating to a data scientist, users now ask natural-language questions that match how they work in the moment, “a user can ask how the artist performed the last time we sold at a venue and get an answer in under a minute," said Brian. "They keep pulling the thread until they have the full picture, which is context that used to take days." It's the kind of insight that helps organizers plan with confidence and deliver a better fan experience.

Because Genie One is built for self-service, the team configured most of the workspace without engineering support, labeling tables and documenting join relationships before developers stepped in to productionize it. "Genie One is something we could figure out and didn't need to loop the dev team in on until the SQL became more complex," said Brian. Genie One queries run through Unity Catalog, so access policies are enforced consistently from the app's service principal down to every catalog, schema, and table, with no separate permissions layer to build or maintain.

Ticketmaster's results with Databricks Genie One: days to under a minute

With Wizard, Ticketmaster turned hours and days of analysis into answers delivered within minutes and impact that reaches beyond speed. Senior front-end developers and source data teams have reclaimed roughly two weeks per quarter that were once consumed by ad hoc requests. The increase in capacity is invaluable, giving engineering more time for the work that makes the fan and organizer experience even better.

More importantly, questions that used to die quietly now get answered. “With Genie One embedded in the flow of work, questions no longer die on the table,” said Brian. “An executive can ask something they never would have submitted as a request and get a confident answer before the moment passes.”

Wizard now serves roughly 400 monthly active users across operations, client relations, communications and senior leadership, with a total addressable base of 700. Across the organization, teams are quickly adopting Wizard and seeing the value. The next phase is already being planned, extending Wizard to every authorized user and adding new tables that deepen descriptive data on events, venues, and markets. The team is also tracking the agent capabilities rolling out in Genie One, which will let users ask higher-order questions like ‘Is there enough fan demand to add another date in this market?’ and have Genie One reason across multiple queries to respond. “Genie One gave us an agentic conversational interface without requiring us to build one ourselves,” said Brian. “It raised our teams’ expectations for what data should deliver, and we’re going to keep raising the bar.”