Product descriptions:
easyJet carried nearly 90 million passengers last year, flying more than 1,200 routes, with a fleet of 300+ planes. Even as one of Europe’s leading low-cost airlines, the company faced mounting pressure to modernise, as traveller expectations necessitate seamless booking, flexible changes and personalised offers. Across Europe, high costs, thin margins and digital-first competitors only raised the stakes for all airlines. With core systems still tied to a decade-old desktop app and one of the region’s largest SQL Server databases, easyJet struggled with frequent engineering bottlenecks, fragmented workflows and more. Moving to the Databricks Platform helped the airline consolidate their commercial tools into a single hub and cut deployment timelines from up to nine months to three — a pace shift that put innovation back on track for easyJet.
Confronting airline pricing pressures from legacy tools
Airline pricing has always been unpredictable, but with changing weather and continually shifting traveller behaviour post-pandemic, it is more complicated than ever before. At the same time, customers expect flexibility when it comes to choosing seats, adding bags or changing bookings on the fly. For easyJet, keeping pace with the industry meant revenue management could no longer be constrained by decade-old technology and fragmented stacks. easyJet wanted to introduce a web-based hub that brought all three steps together. Ultimately, the hub would give trading teams a single place to analyse data, decide on the right fares and ancillaries and act by updating prices or inventory — cutting the lag between insight and execution.
To make this possible, easyJet needed to rebuild their data foundation to better keep up with real-time transactions. Every day, revenue managers adjusted seat fares, updated ancillary prices and logged activity through the pricing interface, creating a steady stream of transactional data. At the same time, analytical data was required for forecasting demand, reporting performance and benchmarking against competitors. Connecting these two layers of intelligence became a critical priority—linking day-to-day pricing actions with long-term commercial planning in ways that weren't possible before. The vision was to enable automated alerts that could flag anomalies, such as a nearly full flight priced too low or a low-traffic route priced too high, catching issues before revenue was lost. In parallel, easyJet recognised that recurring processes tied to revenue management had to become more reliable and less dependent on manual oversight. What had been a fragmented, reactive process needed to evolve into an agile system that could keep pace with modern airline retailing—and the new platform would lay the groundwork to get there.
Struggling to tap into additional market opportunity
Playing a large role in easyJet's strategic priority to strengthen revenue, the push to modernise revenue management began with a daunting set of hurdles. Whilst easyJet's proprietary revenue management system (RMS) is one of the best in the business, it still ran on a decade-old .NET desktop application — a mission-critical tool that hadn't seen a major feature update in years. One of the largest SQL Server instances in Europe supported the app but had gradually been opened up across the company. "Over time, our SQL Server just became a free-for-all. Different teams were connecting directly to pull data, which slowed performance and left us with no real governance and high maintenance costs," said Dennis Michon, Head of Data Product at easyJet. Engineering faced additional strain with 100+ repositories and thousands of lines of code per endpoint, creating data sprawl that further slowed delivery. Worse, migrating functionality often stretched six to nine months and made innovation nearly impossible. Recognising these obstacles, easyJet chose to invest in the Databricks Platform as the foundation for rebuilding their revenue management system.
Simplifying app development to automate manual workloads
First implementing Delta Lake as their analytical backbone, easyJet aimed to replace their legacy SQL Server as the company's main analytical store. By storing both historical and analytical data using ACID reliability, data engineers will now be able to power new capabilities, like demand forecasting, revenue reporting and competitor benchmarking. This will transform analytics from a back-office function into a driver of commercial decision making for business users. Next, easyJet added Lakebase, a fully managed Postgres engine built for the Lakehouse, to their Databricks investment. "Lakebase captures live operational data and turns day-to-day activity into actionable signals for the team. It means our teams aren't just looking at data, they're able to respond to it in ways that directly impact the company's performance," explained Dennis. Feeding this information to downstream systems, Lakebase ensured booking channels remained current and pricing aligned with market realities, making it the new operational heartbeat of easyJet's revenue management operations.
Across the lakehouse, Unity Catalog is being implemented to centralise permissions, auditing and lineage, ensuring that only governed APIs and service principals interact with data. For easyJet, this approach is replacing the cumbersome SQL Server environment—where technical teams often ran direct queries for data discovery—and will improve both performance and compliance by eliminating uncontrolled access. As this governance layer matures, fine-grained control will allow features like Lakeflow Jobs to run even more smoothly. By defining workflows declaratively, Lakeflow Jobs automatically refreshed fares with near real-time latency, generated commercial performance reports and ran scheduled demand and competitor checks. This approach removed the need for engineers to manually monitor pipelines, kept both transactional and analytical layers consistently up to date and increased the reliability of time-sensitive revenue workflows.
With improved governance and more reliable workflows, easyJet tackled data sprawl next. With Lakeflow Declarative Pipelines, teams consolidated 100+ repos with repeatable pipeline specs, streamlining how new data sources were ingested into bronze (raw), silver (refined), and gold (business-ready) layers. Tightly integrated with GitHub through Databricks Repos and deployed with Asset Bundles, these pipelines are also beginning to streamline CI/CD, helping accelerate deployments and cut down release cycles of new apps and app versions. As the team continues to reduce custom code paths, changes become safer, rollbacks faster and maintenance lighter—building momentum for sustained innovation.
At the application layer, easyJet rebuilt revenue management in Databricks Apps, creating two decoupled applications within the environment. The first was a Python API app that connected the lakehouse (analytics) and Lakebase (transactions) through governed endpoints, ensuring operational updates flowed consistently into the system. The second Node-based UI app served as the trading teams' dashboard, where they will be able to analyse demand, compare fares across competitors, decide on strategies and execute changes in one place. Databricks Apps eliminated the need for separate CI/CD and manual infrastructure management, enabling the data product team to deploy business-ready apps faster and iterate directly with users across the easyJet trading team. To extend the efficiency they were already experiencing on the Databricks Platform, easyJet is shifting their business dashboards to Databricks SQL. Rather than relying on legacy SQL Server reports or querying production systems directly, business users are gaining access to dashboards fueled by gold tables in Delta Lake—the same data that feeds both new applications. As this migration progresses, teams are building confidence in the unified data foundation that will support their evolving analytics needs. Together, Databricks Apps and Databricks SQL replaced patchwork tools with a unified, governed hub for revenue management. With this modernised foundation, easyJet positioned themselves to expedite revenue management and also pursue their next vision: agentic RMS powered by Mosaic AI and conversational interfaces.
Modernising revenue management delivers measurable gains
The shift from legacy systems to Databricks quickly paid off for easyJet, largely due to Databricks Apps. Time-to-market for migrating their .NET desktop app to a fully functional Databricks App dropped from roughly 6–9 months to just 3 months—turning a slow, high-risk migration into a fast, repeatable rollout. Apps also removed the complexity behind deployment and CI/CD by using asset bundles to manage resources, permissions, and authentication, simplifying maintenance and scaling. Engineering efficiency improved as 100+ Git repos were consolidated into just two, reducing code sprawl. Governance and security — previously hindered by uncontrolled SQL Server access — are now centralised through Unity Catalog, making role-based access control and compliance easier to enforce.
The business impact extends far beyond the numbers, however. Pricing and trading teams are beginning to work within a centralised, modern hub instead of juggling multiple Tableau dashboards, spreadsheets and legacy apps. As more workflows migrate to this unified environment, the fragmentation that once slowed decision-making is steadily giving way to a more streamlined operational model. With Lakebase feeding real-time transactional updates and Delta Lake powering long-term analytics, decisions happen in near real time and flow directly into airline booking channels. Lakeflow Jobs and Declarative Pipelines eliminated manual oversight of refreshes and reporting, giving engineers back valuable time for new app development. Since Databricks Apps provide an easy-to-use, governed interface where non-technical users can work confidently, the result is greater agility, fewer failures and a lower barrier to innovation.
easyJet sees their recent momentum as only the beginning of their journey. "With our RMS modernisation nearly complete, we're now looking ahead to what's next, like agentic AI assistants to support business users, GenAI conversational interfaces for faster exploration of pricing scenarios and hundreds of new BI dashboards and apps powered by Databricks SQL," elaborated Dennis. As easyJet continues to phase out their legacy SQL Server environment, the platform they're building today is laying the foundation for future capabilities: cutting costs, streamlining operations and eventually enabling a retail-style model where offers, ancillaries and fares can be managed as dynamically as any modern e-commerce platform. When all is said and done, Databricks has given easyJet a platform to not only modernise revenue management but also reimagine the future of airline retailing.