- JetBlue solved its data accessibility problems by building a state-of-the-art modern data stack that includes Fivetran and Snowflake.
- A comprehensive data strategy has given analysts and leaders access to a variety of datasets that span across many workgroups, enabling them to make better decisions.
- The data engineering team can set up new data pipelines in under two minutes, enabling it to focus on more valuable projects.
- JetBlue uses Fivetran’s log-based change data capture to replicate aircraft maintenance data from its on-premise Oracle databases to Snowflake. In addition, the team uses Fivetran’s SaaS connectors to quickly and easily replicate data from third-party systems into Snowflake.
- ELT: Fivetran, dbt Cloud, Azure Data Factory
- Warehouse: Snowflake
- Sources: Oracle, SQL Server, Qualtrics, Servicenow, Jira, Concur, Salesforce
- Cloud Platform: Microsoft Azure
JetBlue carries customers to more than 110 cities throughout the United States, Latin America, the Caribbean, Canada and United Kingdom, with an average of 900+ flights per day. Every person, plane and journey generates data points that reveal customer sentiments, inform revenue forecasting, help predict fuel consumption, prescribe aircraft maintenance and give critical insight into operational readiness.
It all adds up to an enormous amount of data.
“Data is so critical to the airline industry — we couldn't operate without having information at our fingertips to make decisions every single minute of the day,” says Ashley Van Name, JetBlue’s general manager of data engineering. “We're a 24/7 operation, and every minute requires information for our crewmembers to help us get our customers from one place to another efficiently.”
Taming and organizing this overwhelming flood of information was challenging for JetBlue until it created a one-stop shop for its data. JetBlue decided to invest in building a modern, cloud-based data stack. Using pipelining tools like Fivetran, JetBlue is now able to move data from multiple sources to its Snowflake data cloud — allowing Van Name’s team of data engineers to rapidly access information for analytic use cases, and significantly reduce the time it takes to manually build data pipelines.
“We've been able to set up data pipelines in under two minutes,” Van Name says. “The work that would've previously taken engineers weeks, if not months, to fully build, test and deploy, Fivetran makes possible in minutes.”
JetBlue’s Snowflake data warehouse now contains over 115TB of data from 130 different systems, and fresh data is readily available for analysis. The airline has built a suite of self-service analytics products, used by analysts and leaders across many workgroups to deliver meaningful insights. As a result, JetBlue has been able to create better customer experiences and better understand its own operations.
In the next section, we’ll show how JetBlue built a modern data stack to accomplish all that with minimal effort.
Taking airline customer service to new heights
JetBlue generates and stores data in many different business applications and transactional databases. Bringing all of that data into Snowflake quickly and accurately was a challenge.
To centralize that data from the airline’s SaaS platforms and make it accessible for analysis, JetBlue needed a modern data stack built on Fivetran and Snowflake.
The airline’s integration of Qualtrics customer survey data illustrates how simple the process is now. A typical Qualtrics survey asks JetBlue passengers about their in-flight experiences, how easy it was to navigate the airport and how the overall experience could have been better.
To gain a better understanding of its customers, JetBlue uses customer data stored in Snowflake from a wide range of sources to identify survey candidates. Then, a custom pipeline sends information to Qualtrics to distribute surveys. Next, Fivetran’s Qualtrics connector writes customer feedback to Snowflake for analysis.
“Using Fivetran to replicate Qualtrics data to Snowflake is much faster (and easier) than using home-grown pipelines to do the same thing – and we know this, because we tried it,” says Van Name. “It took us a few weeks to get the pipelines working correctly in our home-grown system.”
Consequently, JetBlue’s data teams can take the survey feedback, send it to leaders who can make changes around that feedback and ultimately act on the information to enhance the experiences for customers that choose to fly with JetBlue.
"Fivetran enables us to quickly replicate data from a whole host of source systems into a target database or data warehouse of our choice,” Van Name adds. “This really enables us to break down data silos.”
This strategy has freed the airline’s data engineering team from having to spend time building custom data pipelines. Now, data engineers can instead focus on more valuable projects.
Performing proactive data-driven plane maintenance
For enterprise applications like its aircraft maintenance system, which store data in on-premises data stores, JetBlue uses Fivetran’s log-based change data capture to replicate aircraft maintenance data from its legacy systems to Snowflake.
JetBlue uses this data, whether it's on the ground or in its support centers, to avoid maintenance-related flight delays, proactively solving problems before they happen.
With this data consolidation, JetBlue is getting closer to being able to do truly predictive maintenance, an aspiration for many airlines. In other words, JetBlue will be able to use large data sets to predict maintenance they will need to perform.
With predictive maintenance, JetBlue will be able to improve maintenance efficiency and cost savings while also preventing mechanical issues that could shorten the overall lifespan of a machine.
JetBlue uses data to win customer loyalty and grow its business
JetBlue’s approach to data has helped it maintain a high value on customer experience in the airline industry. And Van Name says that getting data into the hands of people quickly will unlock even more opportunities.
“The more we can do to decrease the amount of time it takes for us to get data to the folks who want it and need it, the more exciting (and effective) data analytics will be,” Van Name says.
With a modern data stack and a data-driven culture, JetBlue is well positioned to grow its business even more in the years to come. That’s an exciting outcome indeed.