Our discovery research began by testing the hypothesis that pharmaceutical research scientists rely on multiple life science data sources to support their workflows. If confirmed, we aimed to understand whether this reliance created a significant enough challenge to warrant deeper investigation and potential solution development.

Recognizing that each scientist involved in drug development faces distinct challenges and data needs, we knew a one-size-fits-all approach wouldn’t work. With a wide range of roles across this complex value chain, we had to prioritize: where, if at all, are scientists struggling most with decentralized data and where, if at all, could we make the greatest impact?

Here’s what we did:

• Leveraged internal subject matter expertise and desk research to identify three key jobs for initial exploration: safety and toxicity, target identification, and drug candidate optimization

• Conducted 9 interviews (3 for each persona) and compared results against each other to assess the best (if any) use case to pursue

  • Note: while it might be untraditional to select a sample size of 3 for qualitative research, this was a calculated risk. As we were under a tight budget and timeline, we knew that it wouldn’t be feasible to conduct 5+ directional interviews per persona. We believed that insights would start triangulating enough after 3 interviews to provide enough of a basis for for use case selection (and they did!)

• Compared insights across personas to assess the most compelling use case

Here were the study goals:

• Determine the workflow of each job

• Validate or invalidate prioritized assumptions to determine ideal use case for concept testing

• Identify key challenges and outcomes across personas to inform which use case experiences the highest level pain and select for alpha development

And here were the results:

• Drug Candidate Optimization and Target Identification were selected as use cases for more comprehensive discovery research