Deliverability degrades quietly. By the time reply rates visibly fall the damage is weeks old and takes longer than that to undo.
A degraded inbox does not just deliver its own mail badly. Sustained poor engagement on a sending domain affects the domain, and recovering a domain takes considerably longer than resting an inbox.
So the fleet is managed by resting inboxes before they cause domain-level harm, accepting a short-term capacity loss to avoid a long-term one. That trade is uncomfortable exactly when volume commitments are tight, which is when it matters most.
Resting inboxes means sending less. Reporting the rest proposal without the resulting shortfall lets a team accept the proposal and then quietly miss its volume for a fortnight without connecting the two. The gap is stated in the same output.
This is the canonical rule set and the on-demand look. Actually resting an inbox, tagging it and detaching it from its campaign so that sending genuinely stops, is a separate execution skill. Keeping the read separate from the action means you can look without committing.
Skills compound. These are the ones we usually install alongside it.