Early discovery often starts with scattered information, limited time, and a lot of assumptions. I wanted to see whether Copilot could help organize the work that happens before validation: competitive research, feature comparisons, early journey hypotheses, and persona hypotheses.

I built a guided workflow in Microsoft Copilot. It asks what is being researched, who the audience is, and what type of output is needed. From there it creates structured working material and a prompt I can bring into Miro to build the board.

The important boundary is that this is not a replacement for user research. The personas and journeys are hypotheses based on the information available at kickoff. They still need to be tested against real customer evidence.

The kickoff flow end to end: choosing a workflow, picking a research type, refining the product or feature, then narrowing in on the audience. Every step is a tap, not a typed answer.

Each workflow resolves into a structured output instead of a pile of notes. Competitive research becomes a feature matrix. Persona work becomes a brief with goals, behaviors, and pain points. Journey work becomes a set of stages and assumptions that can be carried into a workshop or validation session.