From high throughput screening to translational science, Sanofi has had to adapt its IT approach in recent years. *Omics data as well as clinical and patient data have strained traditional enterprise data warehouses not only in terms of volume. Data variability has in fact been the greatest challenge. 80% of data is unstructured. And we at Sanofi reached a point where we could not change our ETL and data models fast enough. In this session, we discuss how we built a new platform to store and extract knowledge from unstructured data. MongoDB was a natural fit. Leveraging metadata and medical ontology, we are now able to provide a means for scientists to tag, share and explore their scientific data and to support the data deluge of an entire translational medicine platform.