Flix Eradicates Data Silos with a Snowflake-Powered Data Mesh Architecture

When it comes to travelling, the journey is often as important as the destination. Leading travel-tech company, Flix, knows this all too well, and has spent more than a decade providing outstanding bus and rail journeys in more than 40 countries across 4 continents. It also knows that when it comes to connecting travelers to transportation options, data is the golden ticket. 

 In fact, data plays a role in every aspect of the company’s offering, from powering its customer-facing app to driving internal operations like marketing, sales, route planning and pricing.

“Flix is data-driven, but I’m not just using a buzzword,” says Jakob Rissmann, Principal PO of FlixMobility Tech at Flix. “We use data for almost every internal decision. And for that we need an easy to use, self-service analytics platform.” 

Over the last decade, the company's growth has dramatically changed its data demands. As a handful of buses and passengers became two modes of transport and millions of customers, its original on-premises data infrastructure could no longer meet demand. The task of physically adding more servers and disks became unsustainable. But that wasn’t the only issue. The company’s data team was inundated with requests from different departments, and working with data from siloed sources meant time to insight was often slow. 

To solve these issues, Flix sought to move to the cloud, where scalability could be instant, pricing would be aligned with usage and data could be shared seamlessly. The goal was to create a data mesh architecture which would allow for distributed data ownership, self-service and more reliable data. Snowflake’s AI Data Cloud provided the foundation.

Today, Flix rolled out Snowflake’s Internal Marketplace to support its data mesh architecture — a secure, centralized space that allows trusted data products to be discovered, shared and accessed across the organization.

Using Snowflake, Flix can now easily share business-ready data products immediately with the relevant internal teams, so they can be used in marketing campaigns, operations and customer service interactions. But it can also combine data that was previously siloed, which brings a new level of insight. 

“At Flix, there are three levels to our service: pre-travel, travel and post-travel,” says Rissmann. “Before, these things were all treated separately. But when you combine data from marketing, journeys and customer feedback, you start to see connections everywhere. That can be really valuable in helping us understand how to improve our offering to customers.” 

This is something that simply wasn’t an option before, but the gains from combined data sources are now a foundational element of the company’s analytics engine. Its data products power initiatives like route planning, demand forecasting and improving internal workflows. 

“This is only possible with a data platform like Snowflake,” Rissmann continues. “Data mesh helps to build your data organization. Listings in the Internal Marketplace make your data products usable. And when you can connect different data products from different stages, and then put those results into a visualization layer like Power BI, you can see things that wouldn’t have surfaced before.”

One of the most valuable benefits of Snowflake is that Flix can now scale resources with demand. Previously, upgrading physical infrastructure would take months, but with Snowflake the company can expand or reduce compute and storage resources instantly, so it’s always equipped to deal with the variability of travel. 

“When our German railway operator or other mobility providers go on strike, we can face high demands within seconds,” says Rissmann. “At that point we need to scale instantly, and for data analytics you can only do that with a cloud platform like Snowflake.”

Snowflake’s instant scalability doesn’t just mean that Flix can cater to customers more easily; it’s also allowed the company to shift the focus of its data teams. With no need to physically manage, maintain and upgrade physical infrastructure, roles like a database administrator are no longer needed. This means the experts once dedicated to keeping the lights on can be redirected to other more value-adding activities. It’s one of many contributing factors that led to an initial cost saving of up to 30% compared to the previous analytics platform setup.

With a single platform that allows easy access to data, teams at Flix can explore new ways of working, incorporating the latest AI and machine learning capabilities into their workflows. 

“We have a growing number of AI use cases,” says Rissmann. “We use many different technologies to power them, and more and more of that is through Snowflake.”

For example, the company uses AI to analyze incoming communications from customers across a range of contact channels. The AI solution then helps to categorize the requests, which enables customer service teams to streamline engagements and match each customer with the right service agent. 

Flix is also evaluating Snowflake Intelligence, which allows users to query data using natural language. This means business users who don’t have a background in SQL can easily access and analyze data products, and see trends and insights through visualizations. It’s a feature that could pave the way for true data democratization throughout the company.

The Snowflake AI Data Cloud has proved popular among Flix’s users, providing a combination of advanced features and intuitive interfaces that allows the company to get to the heart of its data.

“Quite often when you introduce a new tool, there’s initial excitement, but at some point you find the limits of its capabilities,” says Rissmann. “We haven’t really faced that with Snowflake. You can see that it’s a tool made by engineers for engineers — it evolves with the features we need, and considers our feedback and ideas seriously.”

This feedback is delivered through regular calls with Snowflake, where Flix can also keep up with new developments on the platform. The company has already earmarked several upcoming Snowflake capabilities that it plans to use including potential data streaming use cases, and it intends to further explore the potential of AI. 

Ultimately, with Snowflake, Flix has put domain experts from finance, operations, marketing and other key areas in charge of their own data, enabling teams to become more self-sufficient, move faster and make better decisions. It’s a journey that has transformed the way the company uses data, and there are plenty more miles left to explore.