Modernization of Insurance company database systems
Optimizing global portal databases helping retailers and cities to monitor the recycling of 50 billion used bottles and cans a year in the world
Protector
CUSTOMER SPOTLIGHT
Protector offers liability insurance for large and medium-sized companies and the public sector in Norway, Sweden, Denmark, Finland and the UK.
Project Challenge
The customer maintained a suite of over 40 microservices, all backed by a single, self-managed Microsoft SQL Server cluster. While once efficient, this centralized setup had grown into a major bottleneck due to:
- High licensing and infrastructure costs
- Limited scalability and availability
- Difficulties in maintaining and updating the monolithic SQL Server environment
- Inconsistent and degraded developer experience
As a result, performance, agility, and operational efficiency were being compromised. To modernize their stack and adopt a more cloud-native approach, the customer, in collaboration with GCP and CHDS, decided to migrate all microservice databases to Cloud SQL for PostgreSQL.
Solution
We executed the project in well-defined phases to reduce risk, improve transparency, and maintain service continuity throughout the transformation.
1. Discovery and Assessment
We initiated the engagement with a comprehensive discovery phase to understand the technical landscape and prepare a risk-informed migration strategy. This phase included:
- Service-to-database dependency mapping to identify which microservices interacted with which parts of the SQL Server cluster, including shared schemas and tightly coupled components
- Data volume analysis and identification of high-traffic tables and critical paths
- Stored procedure and function audit to evaluate complexity and compatibility with PostgreSQL
- Assessment of schema design issues, such as missing primary keys, inconsistent indexing, and data integrity concerns
- Lifecycle classification of data to identify archival opportunities and reduce unnecessary data migration load
We selected a medium-complexity microservice for a Proof of Concept (PoC) to validate migration tooling, test schema conversion logic, and estimate the effort required for similar services.
This phase was critical for identifying potential blockers early, establishing a migration playbook, and aligning stakeholders on priorities and timelines.
2. Strategy and Preparation
Based on findings from discovery and the PoC, we developed a clear strategy for full migration, including:
- Sequencing the migration based on service criticality, complexity, and interdependencies
- Grooming existing SQL Server databases by removing obsolete indexes, adding missing constraints, and archiving outdated data
- Designing schema mapping from SQL Server to PostgreSQL, including conversion of data types, stored procedures, and indexing strategies
- Defining tooling and automation workflows for schema conversion, data migration, validation, and rollback
All preparation was done with repeatability and scale in mind, ensuring that patterns established could be reused across services.
3. Execution and Migration
Migration was carried out in phases:
- Services were grouped by complexity and risk, starting with simpler, isolated services
- PostgreSQL-compatible schemas were deployed via CI/CD pipelines, with data synchronized and validated
- Application code was updated to support PostgreSQL drivers, queries, and connection handling
- Production cutovers were carefully orchestrated, often using dual-write or blue-green deployment strategies to minimize downtime
We worked closely with the customer’s development and operations teams throughout to ensure a smooth transition for each service.
Outcome
The modernization effort successfully moved the customer from a monolithic, high-cost SQL Server setup to a scalable, cloud-native PostgreSQL architecture on Google Cloud. Key outcomes include:
- Architectural flexibility to support future modernization of business logic, analytics, and microservices
- Significant cost savings by eliminating SQL Server licensing and reducing operational overhead
- Improved scalability and resilience using Cloud SQL’s managed services and high availability features
- Better developer productivity and consistency through simplified tooling and faster feedback loops
- Operational efficiency with automated backups, patching, monitoring, and easier maintenance
This transformation laid the foundation for long-term agility and innovation, aligning the customer’s data platform with modern engineering practices and cloud-first strategy.