Confidence and competence in skills pertaining to data manipulation, analysis, interpretation, and presentation are increasingly sort and valued across numerous sectors but improving these competencies in students can be challenging. There is a disconnect between the steps taken to understand and analyse data and then present findings in a professional setting and students frequently struggle with this. In response to this, I have created fresh resources, introduced new teaching tools, modified assessment strategy and reinforced key concepts in data literacy across the curriculum with an aim of improving student engagement and attainment in the data sciences. This process, as with many advances in improving pedagogical practice, has largely been ad hoc and evidence for initial success and impact began as observational and qualitative. However, long-term cohort wide data sets can come into effect with consistent use of assessment and marking strategy. Here, I show how different evidence can be used to measure impact of pedagogical practice to students and then how this evidence can be applied to guide course curriculums, and more broadly, influence best practice beyond the institution.