With such a tight deadline, traditional research and validation just weren't realistic. Instead, I jumped straight into ideation with Figma Make, prompting it to build the flow and interactions directly, rather than writing documentation or specs first.
This approach let the CEO see and feel the flow interactively, instead of just reading a description or looking at a static wireframe. As a result, the feedback was sharper and more specific from the start, because the CEO was reacting to something real, not an assumption. That meant I could align with the business logic right away, without going through endless feedback cycles.
Because this process was cheap in terms of time, and I wasn't emotionally attached to the first draft, whenever the CEO gave new direction, I could drop the old draft and pivot to the right structure immediately, without losing the sprint's momentum.