Genie has taken data access at BookMyShow from something only a handful of engineers could do to something any teammate across the company can do themselves in seconds. – Noel Curtis, Chief Technology Officer, BookMyShow

 

BookMyShow, India’s leading entertainment technology platform for movie and live event ticketing, sells 22 million tickets a month (264 million annually) to 100+ million monthly active users across movies, live experiences, sports and events. The company was running all its analytics through a single central data team, a bottleneck that limited how fast the business could act. After migrating to the Databricks Data + AI Platform and adopting Unity Catalog and Genie, BookMyShow gave business teams across marketing, finance, live entertainment and cinemas the ability to ask their own data questions in natural language. The shift cut publisher integration times, replaced manual reporting for key partners and reshaped how the company makes decisions.

 

 

BookMyShow’s push toward data democratization began with a clear operational problem: too many business-critical questions were dependent on a small central team. When Noel Curtis, Chief Technology Officer at BookMyShow, joined the company, its data infrastructure was split between a data warehouse on Amazon Redshift and a data lake built on Amazon S3 and Athena, with Glue handling ETL and Apache Iceberg™ as the underlying data format. There was neither a unified data platform nor a unified catalog and governance mechanism. Every analytics request, whether from marketing, the executive team or a business unit like live entertainment, had to be routed through a central data analytics and engineering team.

 

That team fielded the requests, wrote the SQL and delivered results before moving to the next request in the queue. Capacity became the gate on how many analyses the business could perform and how fast. Only power users familiar with SQL could access self-service analytics through Redshift or Athena, and even then, the underlying data wasn’t cataloged in a way that made independent analysis realistic for most teams.

 

“There was practically no self-serve capability,” Noel said. “If you were a power user who understood SQL, you had the capability, maybe, to write SQL queries against Redshift or Athena. But that was really the limiting factor.”

 

BookMyShow moved its warehouse and lakehouse workloads to the Databricks Platform. It also invested early in Unity Catalog to enable data democratization across the company. As Genie became available, that investment paid off faster than expected. “The combination of Unity Catalog and all of the features that Genie provides brings together context with model capability that leads to tremendous outcomes for the organization,” Noel said.

 

 

To scale self-service analytics beyond a handful of technical users, BookMyShow focused on two requirements: making insights accessible in natural language and maintaining consistent governance as usage expands.

 

BookMyShow now runs on the entire Databricks Data + AI Platform, including Unity Catalog, Lakeflow Jobs, Lakeflow Spark Declarative Pipelines and Lakeflow Connect for data engineering and governance, alongside Genie One, Genie Agents and Genie Code. It also uses Databricks Apps and the Databricks Model Router (Unity Gateway), which now handles all of BookMyShow’s LLM traffic regardless of underlying model. Machine learning capabilities in Unity Catalog are also being extended into personalization and search.

 

 

Rather than build a single general-purpose assistant, BookMyShow set up domain-specific Genie Agents for individual business units, including marketing, customer lifecycle management (CLM) and digital marketing, live events and cinema and movie intelligence. Each is grounded in that team’s data context. Genie One remains available company-wide for more general questions, but Noel’s team steers users toward their domain-specific Agents where available.

 

“We find that the outcomes are much better when you’re doing that,” shares Noel. Once teams become comfortable with their Genie Agents, some move on to Genie Code for deeper, more independent analysis and dashboard building, often working alongside BookMyShow’s data engineering team to get there. That marketing and CLM agent alone helps the team track more than 3,500 offers running across payment promotions, promo codes, third-party partnerships and loyalty programs at any given time, work that previously had no dedicated self-service path.

 

Genie One, Agents and Code have now rolled out across finance, live entertainment, movies and cinemas, marketing, sponsorships and brand IPs, with engineering also a heavy user. The questions users ask span the business, from unique transactors and tickets sold for a movie title, to gross merchandise value by city or event category, week-over-week performance of a specific offer type, or top events by revenue over time. “We see very frequently with Genie folks across these departments trying to analyze trends and trying to either diagnose an anomaly or trying to find another theme,” Noel said.

 

 

Underpinning all of it is Unity Catalog, which BookMyShow relies on for fine-grained access control, data classification and lineage, particularly for personally identifiable information, as the company prepares for India’s Digital Personal Data Protection (DPDP) Act.

 

“Something that you get in the data lake ultimately percolates all the way into Genie,” Noel shares. “I never have to be worried about someone from the marketing team asking a Genie question and ultimately sharing a particular type of data asset that would be in breach of some DPDP policies.”

 

 

With governed self-service insights in place thanks to Genie, BookMyShow started applying the platform to high-leverage operational workflows, especially partner reporting and external data sharing. The company’s goal was to reduce manual reporting, speed up integrations and make teams more independent without sacrificing data governance.

 

One of the clearest productivity gains has come from how BookMyShow shares data with its publisher partners, who list its movies and events for lead generation and aggregation. Every one of these integrations used to require custom engineering work. After standardizing on OpenSharing as the single consolidated method for publisher data access, that process now largely comes down to BookMyShow’s engineering team sending an API contract and configuring the relevant filters, cutting integration time from months to about a day.

 

BookMyShow also rebuilt the way it delivers analytics to its Indian Premier League (IPL) ticketing partners. Rather than compiling custom post-game reports for each franchise, the team built a self-service app on Databricks Apps: a single Flask service that queries a serverless SQL warehouse directly, with Chart.js and Leaflet powering the front end and Google Analytics and internal APIs feeding occupancy data. Using Databricks Apps’ built-in authentication, a single data analyst built and deployed the entire app, front end and back end, in a couple of hours, then scaled access from one franchise to four in a matter of days.

 

At a Mumbai Indians game this season, Noel watched the team’s ticketing head pull up live occupancy data on his phone during the match. “I’d never had this before, it’s absolutely a game changer for me,” the ticketing head told him. “It allows me to make decisions at a point in time now before I have to go into the next game, and I can share this data with anyone that I need in my organization.”

 

More broadly, Noel points to a shift in how the organization operates: the central data team now focuses on maintaining Unity Catalog, data pipelines and the medallion architecture as business teams answer day-to-day questions directly in Genie.

 

Meanwhile, the volume of small analytics questions that used to sit in a queue is now largely answered directly through Genie. “Productivity has gone up by an order of magnitude,” Noel said.

 

Looking ahead, BookMyShow expects Databricks to play a growing role in real-time personalization, fraud and bot defense during high-demand ticket sales, natural language customer support and new attribution work for sponsorships and ad tech. Governance remains central to that roadmap as well: “Databricks is going to be the governance backbone for us with Unity Catalog,” Noel said, pointing to the DPDP Act as an ongoing priority.