It does not feel right can be an honest reaction. On its own, however, it does not explain what needs resolving. I try to translate that reaction into an observable issue rather than dismiss it or pass it on unchanged.
Unpack the comment
Imagine a page where the heading feels overwhelming. Before asking for a smaller size, I ask why it is uncomfortable. Does it delay the main content, change the identity’s tone or overflow on mobile? Similar reactions can point towards different solutions.
Once the problem is clear, the designer or writer can propose an answer. Prescribing every detail narrows the space in which their expertise can operate. Giving direction differs from doing the work through someone else’s hands.
Use three connected statements
I describe what I see, its effect on the objective and the change being sought. Four subjects appear in the opening; this delays recognition of the main question; let us foreground one question first. That gives more usable direction than simply asking for simplicity.
I also distinguish observable errors, hypotheses worth testing and personal preferences. They should not all arrive as mandatory corrections.
Own the consolidation
Conflicting team opinions should be resolved before they reach the maker. Two people requesting both brevity and more detail need to agree what matters first. A review meeting should help make that decision.
The approval workflow should establish whose feedback is decisive at each stage. Reopening first-round questions in the final round affects both timing and trust.
Review the revision against the same purpose
I assess the new version using the original objective rather than inventing fresh criteria each time. If the issue is resolved, it may simply take a moment to adjust to a different-looking answer.
Good feedback is not measured by how closely the result resembles my first mental image. It is measured by whether the shared purpose is better served.