Dataomstilling

Dataomstilling is a set of activities aimed at reorganizing and transforming an organization’s data assets to support new information architectures and business processes. It encompasses data migration, data integration, data cleansing, data enrichment, data modeling, and metadata management, with an emphasis on aligning data with strategic goals and governance.

The scope of dataomstilling includes inventorying data sources, mapping data flows, and designing transformation logic that converts existing data to target schemas or semantic models. It often involves moving from legacy systems to modern data platforms, consolidating data from multiple sources, and adjusting data structures to enable analytics, reporting, or operational use. Techniques such as ETL, ELT, and data wrangling are commonly employed, along with decisions about schema-on-write versus schema-on-read and the handling of master data and reference data.

Key components are data quality, metadata management, data lineage, security and privacy, and ongoing governance. Successful dataomstilling requires cross-functional collaboration among data engineers, data architects, data stewards, and business owners, as well as clear standards for data definitions, quality metrics, and access controls. The process is typically iterative, with testing and validation to ensure that transformed data preserves meaning, accuracy, and compliance.

Benefits include improved data interoperability, more reliable analytics, better decision support, and stronger regulatory compliance. Challenges can include data fragmentation, semantic inconsistencies, data quality gaps, security risks, and the need for sustained governance and change management. Related concepts include data migration, data integration, data governance, and data quality management.