A data mesh supports distributed, domain-specific data consumers, with each business domain handling their own data pipelines and data products to fuel self service analytics. It removes the friction for data producers and consumers, hiding complexity, enabling autonomy, and embedding innovation. However, domains cannot truly own their data if they don’t have data observability to have a single pane of glass to continuously view 'known unknowns' as well as 'unknown unknowns’, for example, data quality issues, anomalies, security breaches and privacy violations.