Accounts usually stop scaling because the creative ran out, not because the budget did. Producing at volume without producing filler.
An account plateaus, the budget goes up, and performance gets worse. The usual reading is that the audience is exhausted. More often the creative is, and the two look identical in a dashboard.
Fatigue scales with impressions rather than with the calendar, which is why a monthly content calendar is the wrong unit for paid creative. At low spend a single strong ad can run for months. At high spend the same ad can be finished in a fortnight, because the same people have now seen it eleven times. Any production plan pegged to a date rather than to delivery will be behind at exactly the moment scaling starts working.
Volume is therefore a requirement, and volume is where quality usually collapses into filler. What keeps it from collapsing is treating angles rather than executions as the unit. One product has several genuinely different reasons to buy: the problem it removes, the time it saves, the thing it replaces, the objection it answers. Each of those is a distinct ad rather than a recolour of the last one, and each can then be cut for every placement from the same shoot.
That is also why we plan shoots to produce a month of assets in a day. Not because more is better, but because the alternative is making creative under deadline pressure in the week performance starts dropping, which is when the worst decisions get made.
The unglamorous discipline underneath all of it: kill ads on evidence rather than on taste, keep the winners running until they genuinely decay, and accept that the ad you dislike most is sometimes the one paying for the quarter.