With support from the FOSB infrastructure fund, we started a pilot project to establish a data catalogue which provides an overview of the various databases at INBO, their key statistics and points of contact. This system is meant to be complementary to our CRIS system and aims to increase the findability of our databases for both internal and external users. We also aim to automate harvesting of database information as much as possible: In order to do this, we connected the Open Source data management system ‘CKAN’ to our SQL Server instance. To facilitate the harvesting of information, we produced an ‘extract, transform, load’ (ETL) procedure to centralise metadata stored using extended properties. This poster will detail the technical setup and lessons learned.