You should usually revise it out loud. A case structure is a working hypothesis, not a contract. When new information changes the situation, updating your approach is often a sign of strong judgment rather than a sign that your original structure was bad.

The key is to do it cleanly. Briefly state what changed, explain which branch matters less now, and show how you want to redirect the analysis. That makes the pivot feel intentional rather than reactive.

Candidates sometimes cling to the original structure because they are afraid of looking wrong. In most cases, the bigger risk is ignoring new information just to preserve a framework. You may also find it helpful to read how to update your thinking when new data arrives if you want a more structured way to practice this.

I think saying the adjustment out loud is almost always better. Otherwise the interviewer cannot tell if you noticed it.

This used to throw me off too. Once I started treating the structure as a hypothesis, the pivot felt much more normal.