Migrating from on-premises to cloud infrastructure is a major leap for any organisation. This was the case for a Fortune 500 insurer seeking to modernise its operations. Their existing system, which relied on outdated methods and manual processes, was no longer fit to handle the enormous amounts of data generated daily. This transformation involved building a robust, automated platform, transitioning to the cloud, and optimising for efficiency and scalability. Customer software development company Symfa explains how it supported the firm on this journey.
The insurer’s data management challenges stemmed from diverse inputs arriving from across Europe, in various formats and languages.
Its outdated system struggled to process this data, leaving them reliant on manual Excel workflows for critical tasks like tracking premiums and claims. T
o resolve this, the insurer partnered with Symfa, a development team tasked with creating an automated ETL platform. This platform would stabilise data, ensure traceability, and provide actionable insights for decision-making.
The solution: Automating data processing on-prem
The initial development phase, completed between September and December 2022, involved building an on-premises International Data Warehouse.
This system automated the processing of terabytes of incoming data daily. Files were verified, cleaned, and mapped into two key formats: premiums and claims.
The results were exported into reports and used for reconciliation and accounting, replacing labour-intensive manual processes.
February 2023: A cloud-first approach takes shape
In February 2023, a global cloud provider proposed moving the platform to the cloud, promising improved speed, flexibility, and scalability.
The client decided to explore this option, considering the potential for dynamic resource allocation, a critical feature for businesses managing fluctuating loads. However, this shift required careful planning to optimise costs and ensure seamless migration.
Spring 2023: Proof of concept development
By Spring 2023, the migration process began with a proof of concept (PoC). Collaborating with the cloud vendor, Symfa supported knowledge transfer, documentation, and infrastructure setup.
The PoC, tested on a major contractor’s data, used Azure ETL applications, including Data Factory, Data Lake, and Azure Synapse Analytics. Following the successful demonstration, the team upskilled for cloud-specific roles.
Upskilling for Azure migration
From April to June 2023, Symfa’s team participated in comprehensive training initiatives, including the AZ-900 certification and a hands-on in-house bootcamp.
This preparation ensured they could handle the complexities of Azure infrastructure, which aligned with the client’s Microsoft-based ecosystem.
The roadblocks: Lessons from a misaligned PoC
Challenges arose in July when the vendor’s initial PoC failed to align with the insurer’s business logic.
After a rework, the updated PoC in August met expectations, paving the way for a hybrid team structure. Symfa, the client, and the cloud vendor began working together to finalise the platform.
Finalising the migration
By September 2023, the team completed the on-prem development and transitioned fully to cloud-based operations.
The new cloud system replicates the on-prem infrastructure while leveraging Azure’s advanced capabilities. Data processing now flows through a structured pipeline in Data Lake, organised in bronze, silver, and gold layers, ensuring quality at every stage.
Looking ahead: The impact of cloud migration
The migration brought noticeable speed improvements, but resource efficiency remains under evaluation.
The insurer’s partnership with an award-winning cloud vendor also provided invaluable expertise, enhancing the project’s success. As operations continue to evolve, this case highlights the complexities and rewards of modernising legacy systems.
Copyright © 2024 InsurTech Analyst
