What’s the difference between data as an illusion and data as a pot of gold? As a data scientist, I see data innovations as the result of close collaboration between data experts, managers, and Healthcare professionals, each with their own role: a “data machine.” A machine that only works if it’s set up properly. And well-oiled. And that’s rarely the case! If the machine’s parts don’t fit together, if a step is skipped, or if you start somewhere in the middle? Then things go awry.

There are numerous examples of this. In gastroenterology, for instance, it is a well-known problem that patients respond differently to various treatments because the pathophysiology varies from patient to patient. There is therefore a need for personalized treatment methods. Many startups have capitalized on this opportunity. Some have a clear vision and a concrete, well-developed plan (such as a diary app that provides irritable bowel syndrome patients with immediate, data-driven lifestyle advice). Others rushed to start collecting “promising” big data (microbiome, genetics, blood markers), only to later wonder how—or even if—any valuable insights might ever emerge from it.

Data science is a process that, just like a machine, has a starting point, a processing stage, and an endpoint. The starting point? It begins with an understanding of the organization (what is the institution’s role and added value in the healthcare chain?), an understanding of the patient, and an understanding of the medical field. Understanding the (sometimes latent) needs of patients, specialists, and the organization. Combine this with data expertise, and the ideas will follow naturally. The end point? A robust application, app, platform, or simply an insight that addresses these needs in a sustainable way. The fuel? Vision and connection!