Customer Success Stories

How Inetum helped NGI transform Geosearch into a future-ready digital platform

Geographic information is the foundation of countless digital applications. Governments, businesses, and citizens use this data every day for maps, location-based services, spatial analysis, and public services. To make this information easy to search and reliable to use, the National Geographic Institute (NGI) developed Geosearch, a free search service that provides access to geographic data through a web viewer and API.

Together with Inetum, NGI modernized Geosearch while also laying the foundation for a new, future-oriented approach to developing, managing, and deploying digital applications.

Publié le: 26/10/2026

From search service to strategic platform

Geosearch brings together eight authoritative data sources in a single, automatically updated environment. In addition to BeST Address, the service includes data on municipalities and streets, place names, cadastral parcels, kilometer markers, and estimated building heights. Users can search using free-text queries without requiring a fully structured address or complex search syntax. The service also supports multilingual addresses, reverse geocoding, and the processing of large address datasets.

When NGI decided to modernize the application to sustain and expand its services over the long term, it became clear that the underlying technical foundation also needed to evolve. Together with Inetum, the organization moved from manual deployments to an automated platform based on container technology, Kubernetes, and GitOps.

As a result, Geosearch became the first containerized product within NGI and the first application to leverage a fully automated CI/CD and GitOps delivery chain. The project served as a pilot for a new technical approach while establishing a foundation for future digital services.

From manual operations to a standardized approach

NGI's existing deployment model had supported the organization for years but was gradually reaching its limits. Releases were manually deployed as WAR files to individually managed Tomcat servers, while server setup and configuration also relied on manual processes.

This approach made deployments time-consuming and environments difficult to reproduce. Because configurations and changes were managed separately for each environment, there was a risk of discrepancies between tested and production versions. Operational knowledge was also partly tied to specific systems and configurations.

NGI wanted to reduce its dependence on manual processes while establishing an approach that could be reused for future applications. The objective was clear: make releases more consistent and predictable, reduce operational risks, and integrate quality and security controls directly into the development and deployment process.

A containerized application and automated delivery pipeline

NGI contributed its expertise in geographic data and search algorithms, while Inetum supported the modernization of the application and deployment architecture. Together, both teams built an automated delivery pipeline that supports the full journey from source code to production.

Geosearch was packaged as a container image and deployed on K3s, a lightweight Kubernetes distribution. This choice aligned with the size of the team, the number of applications, and the expected release frequency.

Developers manage their code in GitLab. Automated pipelines build container images and perform quality and security checks. Changes are then deployed to the different environments through GitOps-controlled processes.

A key principle is that the application is built once and the same tested version moves through every environment. Environment-specific configurations are managed separately. This makes deployments more consistent and reduces the risk of differences between test and production environments.

Deployments no longer depend on manual server actions but follow a standardized and auditable process. Changes remain traceable, and controls are integrated into the delivery pipeline rather than being performed only afterward.

A reusable platform for future applications

The modernization went beyond the application itself. NGI maintains separate environments for development, integration, user acceptance testing (UAT), and production. Through infrastructure automation and GitOps, both infrastructure and applications are managed consistently.

Security and manageability were also built into the design. The architecture provides a clear separation between internal and internet-facing network zones, controlled access mechanisms, and encrypted storage of application secrets. Configurations, certificates, and changes are managed through auditable processes.

In addition, a centralized observability environment was implemented, enabling developers and administrators to access metrics, logs, and traces across multiple environments. This shared operational view supports release monitoring, incident investigation, and platform health management.

The combination of standardized deployments, centralized visibility, and automated operations provides NGI with a foundation for developing and deploying future applications according to the same principles.

The architecture also emphasizes open technologies, including Apache Solr and K3s, as well as open data exchange formats such as JSON, GeoJSON, and XML. This supports Geosearch interoperability and can help reduce dependence on specific vendors, although technological autonomy also depends on broader infrastructure and operational choices.

A collaboration that builds knowledge

The collaboration combined NGI's expertise in geographic information and search algorithms with Inetum's experience in application modernization, container platforms, and DevOps practices.

By working closely together, both teams not only delivered a modernized application and deployment architecture but also built valuable knowledge for future modernization initiatives. NGI teams were trained, and the new approach was fully documented.

Today, this approach has become the standard for future deployments across the organization. Geosearch serves as a first practical example of how NGI can manage applications in a standardized and automated way.

A foundation for reliable digital services

With the modernization of Geosearch, NGI has taken an important step in renewing its technology landscape. The organization now benefits from a containerized application and an automated deployment pipeline that support greater standardization, control, and reproducibility.

The result includes:

A containerized application architecture based on K3s.
Automated pipelines for building and validating applications.
Controlled and traceable deployments across environments.
Centralized visibility into metrics, logs, and traces.
A documented and reusable approach for future applications.

The greatest value lies not only in the modernization of Geosearch itself. The project enabled NGI to validate a new approach to application management and deployment in a real-world setting while building the expertise needed to apply that approach more broadly.

At the same time, Geosearch continues to support the accessibility and reuse of authoritative geographic data in Belgium. Public organizations can integrate the API free of charge or, through collaboration with NGI, establish their own instance with additional datasets or custom developments. The service promotes interoperability and offers an alternative to organizations that would otherwise need to develop a similar search solution themselves.

Through Geosearch, NGI combines two ambitions: making geographic information accessible and reusable while building a technical foundation that can support future digital services.

Request a contact