With the arrival of web mapping and spatial data warehousing, using geospatial data has become common. In fact, it is widely accepted that 80% of all data are spatially referenced. However, when one starts integrating geospatial data from different sources, numerous hidden problems appear and data quality maybe questioned. It is only when one knows about the underlying nature of geospatial data and about their inherent heterogeneity that one can find solutions. This presentation will focus on the challenges facing Spatial ETL processes (Extract, Transform, Load) when doing spatial data warehousing, spatial on-line analytical processing (SOLAP) and spatial dashboard. In particular, it will focus on the high level of diversity that exists when doing spatial referencing (ex. positioning), when measuring geographic phenomena (shape, size, orientation), when describing them in computers, and when using them. Facilitating solutions (ex. standards, Spatial Data Infrastructures, technologies, data quality awareness) and the role of Geomatics knowledge to make the right choices when building a system will be emphasized.