AP Pension revolutionizes its data governance and analytics with Microsoft Fabric

Planning for retirement is both financial and deeply personal. A pension is not just about the profitability of your savings, but also about the value of the investments for society and the funds available for you as a retiree.

As a customer-owned company based in Denmark, AP Pension, can make free and independent choices in the marketplace, focusing on responsible, environmentally sound investments with the best return for customers now and in the future. Beginning as a pension company for a wide range of Denmark’s cooperatives in 1919, AP Pension has continued to evolve to meet the changing Danish society and labor market. In its 100-plus years of service, AP Pension has remained true to its principles of ownership and democracy.

AP Pension had decades worth of line-of-business data from disparate sources and connections. The company's data management and operating-system processes were siloed and burdened with technical debt, hindering the efficiency and flexibility needed in a rapidly evolving financial landscape. The firm was on a cloud migration journey. However, the setup did not have the desired flexibility or efficiency to meet the needs of their users as the company’s demands and offerings evolved.

Jacob Rønnow Jensen, Head of Data Platform at AP Pension, says, "Our wealth of data, rather than being an asset, became a challenge. It was clear; we needed to rethink our systems. We needed a consolidated and democratized approach to analytical data in accordance with our principles and digital strategy.“ AP Pension was seeking a modern platform that could handle all kinds of data, enabling it to move and process that data efficiently and flexibly across different sources and tools.

A future-ready data analytics solution

The search for a solution led AP Pension to Microsoft Fabric. Fabric is a cutting-edge platform that redefines data analytics. Though a relative newcomer to AP Pension, Jensen is a tech veteran who was already familiar with Fabric. He says, “Being part of the private preview for Fabric has been like having a crystal ball to tell me how the future is going to look—and how to change our data strategy accordingly.”

AP Pension found compelling reasons to adopt Fabric; it offers streamlined data management, enhanced scalability and efficiency, and the ability to perform intricate data analytics. Fabric comes fully integrated as a software-as-a-service offering and provides a consolidated view of the firm’s diverse data landscape, bolstering a unified approach to data analysis. It handled extract, transform, and load (ETL) processes more effectively, scaling operations to meet growing demands while delivering optimal performance. And Fabric supported the firm’s strategic decision-making with its extensive analytics capabilities.

“The architecture of Microsoft Fabric was more than an upgrade for us,” Jensen says. “Its ability to bring compute to data, rather than the other way around, means we can process information more rapidly and with greater precision. We were finally able to see the future of data-driven decision making.”

Microsoft AI Cloud Partner Program member twoday assisted AP Pension in their Fabric proof of concept. This setup aims to streamline the data workflow, making it easier for business analysts and data engineers to work with data products using familiar SQL queries and business intelligence tools.

Introducing Fabric will have a profound impact on user groups within AP Pension as actuarial teams, advanced users, and citizen developers benefit from the enhanced capabilities and flexibility offered by the platform. The ability to co-create and prototype data products in a secure and collaborative environment facilitates innovation and accelerates the development of solutions tailored to the firm's needs. Jensen says, “We are in a regulated industry, and we are big enough to have the need for an enterprise data platform with automated data management and a centralized approach to the governance, security, and curation of data. But, at the same time, we are small enough that we only operate in one region, fairly close to our users and source systems, and we can make quick and agile decisions. My team manages all the General Data Protection Regulation or personally identifiable information data. In Fabric we anonymize the data at ingestion time so our stakeholders can proceed without worrying about compliance. This frees up domain experts to work in their areas of expertise.”

As an early adopter, Jensen and AP Pension had considerable interest in providing feedback and closely monitoring the evolution of Fabric. Jensen was particularly enthusiastic about Mirroring in Fabric, a feature that can be set up with a few clicks to choose the database you want to mirror in OneLake and then set up a connection. This functionality enables seamless data integration from supported databases into OneLake in Fabric without the complexities traditionally associated with ETL processes. In addition, any data brought into OneLake via Mirroring is exempt from storage costs.

As Jensen says, “Mirroring eliminates the “E” and “L” of “ETL” letting us focus on the “T”.” By simplifying data ingestion, Mirroring empowers AP Pension to efficiently manage its data landscape, ensuring timely and accurate data availability for analysis and decision-making. This innovative feature represents a leap forward in data management, offering a streamlined approach that enhances operational efficiency, cost management, and data governance.”

AP Pension has also been able to completely transform its data delivery by leveraging Microsoft Fabric to implement a centralized medallion architecture with the possibility of workload isolation at the edge. To this end, the organization developed a framework using Fabric APIs to automate the creation of essential Fabric artifacts for data delivery, made up of workspaces, lakehouses, warehouses, and shortcuts.

To manage metadata for over 40 workspaces and hundreds of shortcuts across development, test, and production environments, AP Pension needed an intuitive and secure interface. As an early adopter of SQL databases in Fabric and accoMasterdata—a custom visual enabling writeback from Power BI to Fabric—AP Pension combined these technologies to create a powerful tool for translytical (transactional + analytical) data delivery.

This solution combines writeback from Power BI with the Fabric SQL Database to enable real-time data updates in a fast, transaction-capable engine synchronized with both the SQL analytics endpoint and the semantic model and running on the Microsoft backbone network with built in security without the need for private endpoints.

“When maintaining metadata for your Enterprise Data Fabric, writeback from Power BI and Fabric SQL Databases is a match made in heaven. Having a Power BI custom visual like accoMasterdata with built-in functionality for data entry, data validation, audit trails, and approvals takes this translytical relationship to the next level” Jensen commented.

With the implementation of Microsoft Fabric, AP Pension is able to deliver a comprehensive economic analysis that meets the needs of the organization's superusers. Additionally, the improved scalability and efficiency of the solution have allowed for effective processes. Most importantly, the enhanced data capabilities facilitate a thorough analysis of customer, financial, and investment data, thereby contributing to the firm's ability to fulfill its core mission of providing a secure retirement to its customers.

What’s next

Microsoft Fabric promises to let AP Pension achieve a previously unattainable level of data organization, making information more accessible and meaningful. The efficiency gains will allow AP Pension to scale its operations dynamically, addressing the needs of its customers more effectively. The platform’s analytical capabilities will empower the firm to forecast trends and plan strategies with new insights and accuracy.

The journey with Fabric is just beginning for AP Pension. The organization is keen to explore additional out-of-the-box capabilities like data quality, data catalog, and Copilots and AI to enhance its services and operational efficiency. In short, as Fabric continues to evolve, AP Pension will evolve with it and will continue to test new functionalities and enhancements as they become available.

“The essence of our mission as a company is to secure a bright and sustainable future for our customers,” says Jensen “Fabric supports this mission by bringing internal and external data together in an analytical platform that out-of-the-box accelerates collaboration and data-driven innovation better than anything we have seen before.”

Find out more about AP Pension on LinkedIn.