Scaling on ROAS alone means scaling on whatever Meta chooses to attribute. Two campaigns can report the same return and deliver very different profit once cost of goods, shipping, payment fees and discounts are accounted for.
Define your break-even before you write a rule
Start with contribution margin per order: selling price minus product cost, fulfilment, transaction fees and expected returns. That gives you the break-even ad spend per order, and therefore the honest threshold your campaigns need to beat.
Turn thresholds into automated actions
With a real threshold, automation becomes straightforward. Rules run on a schedule and act on evidence rather than on how the account felt this morning.
- Scale up in controlled increments when spend clears your margin threshold over a meaningful window, not a single day.
- Reduce budget rather than pausing outright when performance drifts, so you keep history and delivery.
- Pause on hard floors: enough spend, no purchases, clear negative contribution.
- Require a minimum spend and lookback window on every rule so noise cannot trigger it.
Guard against automation whiplash
Rules that fire too often teach an account to panic. We cap how frequently budgets can move, stagger checks across the day, and keep a single source of truth for margin so the ad account and the P&L do not disagree.
Why this matters more as you grow
At low spend, manual checks work. At scale, the number of decisions per day exceeds what any operator can make well. Automation built on margin keeps the discipline consistent while spend grows, which is what keeps growth profitable rather than merely large.