By leveraging the efficiency and scalability of Matillion ETL, TUI Group – the world’s leading tourism company – can now deliver more data insights to the business faster. Kai Wolffram, Head of Analytics, TUI Group, discusses how analytics are shaping business transformation in the digital age and the benefits of working with Matillion to develop a more modern solution with a centralised, cloud-based data warehouse that included data from across all hotel brands.
TUI Group, based in Hanover, is the largest integrated tourism group worldwide, operating six airlines across Europe, 380 hotels in more than 30 destinations and approximately 150 medium and long-haul aircraft. TUI serves more than 27 million customers in 180 destinations around the world and employs over 70,000 people.
The challenge
Migrating to the cloud for faster analytics and reporting
As the largest tour operator and leisure company in the world, TUI wanted to use vast amounts of data to glean answers to important questions about business performance. But its data warehouse architecture was limited to one on-premises system supporting only one of its hotel brands. In addition to being a partial data source, the existing system had long development cycles, in which it took weeks for developers to attach new data sources to the data warehouse and build complex business logic for analytics. TUI needed a more modern solution with a centralised, cloud-based data warehouse that included data from across all hotel brands.
The solution
Using the power of the cloud for data transformation
TUI sought a cloud-native solution that would enable an ultra-scalable, fast, simple and cost-effective modern data analytics tech stack. The company also needed a powerful ETL solution that could:
• Consolidate multiple data sources, including Google Analytics, Salesforce and third-party data such as currency exchange information
• Orchestrate and schedule data loading and transformation for different source systems
• Implement complex business logic and create parameterised, reusable shared jobs
TUI selected Matillion ETL for Snowflake to help it move to the cloud, aggregate data sources and provide a flexible platform for data warehouse automation.
“Matillion’s scalability and efficiency were a game-changer for our growing, global business,” said Kai Wolffram, Head of Analytics, TUI Group. “We immediately leveraged the power of the cloud and saw instant value from increased performance. In Matillion, we found a modern, flexible solution that enables true data insights and is fun to work with because it’s so easy to use.”
The benefits
Leveraging data to keep up with the pace of business
Once TUI selected Snowflake and Matillion, the company was able to move initial datasets into the Snowflake cloud data warehouse within minutes. Matillion ETL’s built-in features and native technologies allowed TUI to have its first jobs in test within a couple of hours and running within one week. Matillion’s pre-built connectors allowed TUI developers to create data pipelines to multiple data sources quickly and easily.
TUI finds Matillion’s ELT approach (vs ETL) to be easy-to-maintain, expandable and stable, with a scalable data transformation platform for its cloud data warehouse solution. Since implementing Snowflake and Matillion, TUI has enabled teams across the business to answer more questions with data, faster, and have more confidence in the data behind those answers.
What’s next?
Using Matillion to improve insight across the whole business
To increase access to and use of data to enable intelligent business decisions, TUI plans to move all its systems, across all business units, to the cloud. With this expanded move to the cloud and to cloud-based data transformation, TUI hopes to achieve the same level of speed and insight across its other analytics departments.
Kai Wolffram, Head of Analytics, TUI Group, discusses the key business priorities when working with Matillion and the overall benefits.
Can you provide an overview of your role and responsibilities at TUI Group?
TUIis the world’s largest travel organisation and covers all aspects of the tourism industry so operating on aeroplanes, cruise ships, retail stores and hotels. My role is to take care of all the IT infrastructure in all our hotels and to provide analytics solutions with our analytics competence centre across all hotels and brands – so providing a holistic overview of booking figures and all other aspects that are of interest for our hotels.
How does technology play a part in enabling the organisation to remain the largest integrated tourism group worldwide?
Technology is one of the most critical assets in our industry. The industry has changed quite a bit and is continually transforming. When we started there were no booking systems or Internet booking platforms. It’s a very diverse market and for us, technology is key to providing the best guest experience because that’s what it’s all about. It starts with the booking of your vacation which needs to be seamless, and continues during the journey to your destination and of course when you’re in that destination. Important factors include a good Wi-Fi experience in our resorts, app usage so that guests can get the latest information about the destination and about the properties they are staying in, daily programmes, booking amenities, room upgrades, etc. So there’s a lot of technology and applications involved that we are integrating which all come together to provide the best guest experience possible.
As Head of Analytics, what were some of your key priorities prior to the commencement of the collaboration with Matillion – what were you hoping to achieve?
When I started the role almost five years ago, we didn’t have a centralised data warehouse. The portfolio of hotels was already big, but there was a clear strategy to scale that portfolio, so to add more properties to our journey. There was also a high demand for data and not just booking-related data or guest-related data or financial figures. From a headquarter point of view, overseeing the overall performance of all properties, but also from a local point of view, distributed systems, no standards of booking tools, so it was very hard to have a clear view of how the properties were performing.
Our goal was to create a holistic overview across our portfolio and across all brands. Some of the brands had similar solutions like a data warehouse, some newer brands didn’t have solutions at all and they were not integrated at all. So that was fundamental in our roadmap and that’s where it all began.
How are analytics shaping business transformation in the digital age and how are you leveraging this to keep up with the pace of business?
Technology is key and so is data. Hospitality is a very sensitive ‘product’ – it centres around people and the service we are providing to them, but the baseline is data. We are interested in providing a good overview of data to inform how the properties are performing, what the occupancy is like, available rooms, room rates, etc. We provide people with a lot of useful tools to operate their property and make decisions.
What can you tell us about TUI’s goal to create a centralised data warehouse across all brands and how seamless was this?
When we started, we didn’t need to migrate anything. We were able to start from scratch and that’s what we basically did. So we started a greenfield project and tackled it from a business perspective – we considered what the business needed and focused on what we wanted to achieve first. We also tackled it from a technology perspective – so what stack were we going to use. TUI already has a very strong footprint in the cloud so it was clear to us that everything we would need to build would be in the cloud. So we engaged with Snowflake as our core analytics layer and core data warehouse in the cloud. Snowflake and Matillion built the core of that layer. Everything is hosted in the AWS Cloud in Frankfurt, managed completely centrally.
How did you harness the cloud for data transformation and what benefits has this provided you with?
Cloud is everywhere. Some of our data sources are already in the cloud but a lot of our data sources are still on-premise in the respective destinations. But we are using the cloud to be able to scale because especially in our area of business there are certain areas where the demand is high and certain areas of the day where the demand is not too high. So of course, particularly in the morning, everybody needs their latest reports, the chefs are checking how many guests they have in the property and how many meals they need to prepare etc. There’s a lot happening in the first four or five hours of the day and then demand drops. So that’s a good chance for us to scale our environments based on the usage which is something you cannot do when you are on-premise because then you need to pay for the hardware, whether it’s used or not.
It’s no secret that TUI was hit very hard by the pandemic so particularly during this time we needed to focus on cost. Everybody talks about scaling and of course the business needs to grow, but in certain areas you might also need to scale down if the business requires it.
