What would you do if your MD rejected your model?

First thing I'd do is figure out which version of "rejected" I'm dealing with — they're three different problems with three different responses.

Version one: the MD thinks I made a mechanical error. The right move here is to ask specifically what they're seeing — "can you point me to the cell or the assumption?" — because the worst outcome is that I rebuild the entire model when the issue is one input. If they can't point to a specific number, I'll send them back the three or four assumptions I think are most likely to be the issue, with my reasoning, and ask which one they want me to revisit.

Version two: they disagree with an assumption — say, my growth rate or my exit multiple. This is the most common case and it's not actually a rejection of the model, it's a rejection of an input. I'd ask what they think the right number is and why, then re-run with their assumption AND my assumption side-by-side, so they can see the sensitivity. I want them to feel ownership of the inputs, but I also don't want to silently bury work I think is right.

Version three: they think the model's structure or approach is wrong. This is where junior bankers get into trouble by debating. I'd absorb the feedback, ask one or two clarifying questions to make sure I understand the new direction, and rebuild. I might disagree, but the MD has seen 100 deals to my 1, and the prior is heavily on their side.

Underneath all three: I'd never push back in the moment. I'd take notes, go back to my desk, think about it, and come back with the rebuilt or re-run version plus a written rationale if I still disagreed.