The current paradigm in geospatial data quality assessment is datacentric (internal quality) and the evaluation of the GDEMs does not escape this situation. Data quality should be evaluated as fitness for use, but this perspective is unapproachable. A new paradigm is proposed to overcome this situation by considering generic use cases that link geospatial data with its processing (algorithms). The new approach proposed by the functional quality supposes an intermediate situation between the user’s and producer’s perspectives (external and internal quality). This paper defines the functional quality, explains the need for this new perspective and shows an example.