SWICA moves to the cloud with Azure Synapse Analytics and Power BI

SWICA is one of Switzerland’s biggest health and accident insurance companies. The organization insures a network of over 1.5 million people, handling a premium volume of over 5 billion Swiss francs. SWICA is focused on bringing the best quality of service and ensuring customers are happy and feel well supported. SWICA’s pragmatic and focused approach on satisfaction has brought in the highest satisfaction ratings amongst customers in Switzerland. 

SWICA’s culture is founded on finding new ways to bring more value to customers. A key process in which the organization does this is by leveraging data for accessible intelligence. With most of SWICA’s data on-premises, scaling and adjusting for growth proved a challenge. Fabian Ringwald, Chief Information Officer at SWICA, saw cloud-based data warehousing as an opportunity to take SWICA to the next level. 

Because SWICA deals with highly sensitive data, centralized data and analytics can be difficult. While collaboration and compliance are cumbersome in a legacy platform, the decentralized nature of its legacy systems and architecture fit the internal needs of SWICA. The team wanted to find a middle ground that combined the compliance and security the decentralized system brought with the efficiency and collaboration made possible through a cloud-based solution. The solution needed to be a unified platform with a common analytical landscape that would preserve its current strengths and eliminate pain points that hinder growth and limit opportunities for innovation. 

With Microsoft Azure Synapse Analytics, and Power BI, SWICA built a vision for a unified platform that allows them to leverage sensitive, critical data, in an efficient, safe, and cost-effective way. 

Starting the migration to the cloud

In 2021 SWICA’s data and analytics team, under the leadership of Ringwald, began its cloud migration journey. Microsoft Switzerland and its two Swiss Azure regions allowed SWICA with specific data-residency and compliance needs, to keep its data and applications close. While the team evaluated multiple products, a solution built on Azure Synapse Analytics stood out as the best fit. With Azure Synapse Analytics, SWICA was able to have multiprogram language support, as well as deep integration with Azure services and Spark clusters. Azure Synapse Analytics would support the team’s goals of query speed and show improvements over its current data processes—ultimately enabling better decision making. In addition, a single UI of Synapse via Synapse Studio provided uniformity and consistency across the organization. 

Because Power BI was already in use, Azure Synapse Analytics was able to seamlessly integrate into its existing reporting infrastructure. Azure Synapse Analytics also enabled a broader use of Power BI capabilities. With Power BI, SWICA was able to provide technical business users with the ability to build reports with guaranteed data quality across the organization. 

Using Azure Synapse and Power BI to support a decentralized architecture

With Azure Synapse Analytics and Power BI selected for the unified platform solution, Tobias Rist, Data Solution Architect, began the project. Since SWICA’s new architecture required decentralization to alleviate bottlenecks in data preparation and data engineering, the organization began by breaking down centralized BI teams and monolithic pipelines to create individual data products. While the data products function as independent items that have a single owner and pipeline, products are designed for reusability within the existing architecture. Each of these data components contains its own ETL pipeline: data starts in an Azure Data Lake and is then processed into Azure Synapse Analytics via Spark Engines. Each component has an output port that functions as an API to connect data into downstream data products and reports, including Power BI data models.

The current architecture assigns different workbenches to different domains within the company, and using platform management tools, domain owners can provision shells inside workbenches and start managing data. This enables users to figure out how best to use the data. Users can transform, aggregate, and enrich data as part of its individual data products until they decide how data is best utilized. 

The flexibility of a data mesh architecture allows SWICA to retain efficiency and foster collaboration while keeping data secure and decentralized. With workbench owners working with the data community within SWICA to determine data compliance rules, SWICA can ensure that each data product does not violate any compliance requirements and is properly documented as part of its platform management system. 

Microsoft Purview serves as SWICA’s data catalog. Microsoft Purview makes it seamless for users to search for specific data assets across the company and expose it in the catalog. Users are publishing their metadata to the data catalog to also make data more accessible and collaborative. 

Shifting to Power BI Premium and enabling self-service

With Power BI being used prior to SWICA’s cloud migration, there was a sizable community using Power BI report server. However, with the migration to the cloud, SWICA is hoping to also move its users to Power BI Premium that will open the door for new features and scalability. "With Power BI, even the technical-loving business user is able to build their reports without waiting weeks or months for the figures they’re looking for," says Phillipp Frenzel, Data Engineer at SWICA.

Originally, the SWICA Data and Analytics team oversaw building most of the dashboards and reports across the organization. However, over the past few years, the team has been overwhelmed with requests. To help with the request overload, Azure Synapse Analytics feeds into centralized Power BI datasets across various workbenches that business users leverage to build dashboards. To further education across the organization, the Data and Analytics team is also working to build a center of excellence and help consult users on best practices and help educate on how to effectively use the product.

Bringing the cloud vision to life

While the team has mapped out a migration plan and decided on architectural design, SWICA’s initiative to move its servers to the cloud is in the early stages. The data and analytics team has completed the build of two MVPs and is currently working on a third before starting efforts to move off its legacy platform and build out the architectural foundation. Within the next two years, the revamped decentralized architecture is set to be up and running. Once the platform is fully built out, the team hopes to support more advanced cases and introduces Artificial Intelligence to elevate data and visualizations. With more advanced visualizations, SWICA can provide a better quality of service, ensuring SWICA customers feel well supported and happy. 

Individuals across SWICA are enthusiastic about the new architecture and curious to see the opportunities and scalability that will arrive as a result. The data and analytics team are already beginning to educate users on how the system will change and how this new platform with Azure Synapse Analytics and Power BI will make data transformations and migrations even easier across the organization. With its sights set on the cloud, SWICA is excited about the new platform and the growth and innovation it will bring to the data and analytics pillar of its organization.