Scale Up Fast, Scale Down Slow: The Asymmetric Budget Rule That Saves Ad Accounts

The day you panic and slash your budget is usually the day you turn a temporary dip into a real one.

Here's the cost. Your account has a bad day. Cost per acquisition spikes, ROAS drops, and your gut says do something. So you cut the budget hard, maybe halve it. The next morning performance is worse, not better, so you cut again. Within a week you've gone from spending properly to barely spending at all, your best campaign has lost its learning, and the recovery that would've happened on its own now takes you a month of rebuilding to claw back.

I've watched founders do this to accounts that were completely fine. The dip wasn't the problem. The reaction to the dip was.

So I want to walk through the single rule that's saved more accounts for us than any clever bid strategy: scale up fast, scale down slow. The two directions are not symmetrical, and treating them like they are is how good accounts get wrecked.

Why up and down move at different speeds

The instinct most people have is that scaling is one dial. Performance good, turn it up. Performance bad, turn it down. Same speed, same logic, just opposite directions.

In reality, the account doesn't behave symmetrically, so your hand on the dial shouldn't either.

When you increase spend, the change is yours to make and the system absorbs it reasonably quickly. When you decrease spend in a panic, you're not just spending less. You're yanking budget out from under a campaign mid-flight, resetting what it's learned, and very often making the exact number you were worried about even worse. Up is a decision. Down, done badly, is a wound.

So we run them on two completely different clocks.

Scaling up: lean on yesterday, move 20% at a time

When the account is hitting its target, scaling up is the easy half. I look at one thing: yesterday.

If yesterday's new-customer cost per acquisition came in at or below target, I increase the budget by 20%. That's it. Not double, not a guess, just a steady 20% step on the campaign that's working.

And I deliberately ignore today's numbers while I'm doing it. People tie themselves in knots here. "Yesterday was great so I should scale, but it's 2pm and today's cost per acquisition looks high, so I'll hold off." No. Today is half-baked, the day isn't finished, and intraday numbers swing around for reasons that have nothing to do with your decision. I make the call off the last complete 24 hours and I make it calmly. Hit target, up 20%, move on.

The only time I break the 20% rhythm and scale faster is a genuine short window where you have to spend, like a Black Friday weekend or a three-day promo. Then you push hard because the clock is the constraint. Outside of those moments, surfing the budget up and down multiple times a day has never given me consistent results. The slow, boring 20% step has.

Scaling down: the falling knife

Here's where the discipline actually matters, because here's where the damage happens.

When performance dips, the temptation is to cut immediately. Don't. Cutting into a bad day is catching a falling knife. You reach out to grab it, and every action you take while it's falling just does more harm. You cut, it gets worse, you cut again, it gets worse again, and now you're three moves deep into making a temporary wobble permanent.

What I've found over years of doing this is almost annoyingly simple: when you have a bad day, do nothing. Don't touch the account. Leave it alone for two to three days and, more often than you'd believe, it auto-corrects on its own. The spike settles, the cost per acquisition drifts back toward target, and the thing you were about to amputate turns out to have just needed a day or two to breathe.

That's the asymmetry in one line. I'll scale up off a single good day. I won't scale down off a single bad one.

The three-day no-touch window

So the rule I actually hold myself to: before I cut a budget for poor performance, I want three to four straight days of the cost per acquisition sitting above target. Not one ugly day. A real, sustained run.

One bad day buys you nothing but a no-touch window. Three or four bad days in a row is a different signal, and only then do I step the budget down, and even then only by that same 20%.

I'll be honest about why this is hard, because it's not the logic that trips people up, it's the feeling. Sitting on your hands through a bad day looks like you're being lazy. If you're running this for a client, doing nothing while the numbers look rough can feel like you're asleep at the wheel. I've felt that. But I'd rather look idle for 48 hours and let the account recover than look busy and cut the legs out from under a campaign that was about to be fine. The success rate on patience is just much higher. That's not a vibe, it's what the accounts keep showing me.

The no-touch window helps on the way up, too. When you scale up and the cost per acquisition rises a bit in response, don't immediately scale back down. Hold it at the new budget and let it settle. Sometimes it stabilises back to target at the higher spend and you can step up again. Sometimes it doesn't, and that's the account telling you you've hit a ceiling and the real fix is new creative, not a smaller budget. Either way, wait and read it, don't flinch.

The floor that makes patience safe

Now, "do nothing and wait" only works if you've given yourself a floor. Otherwise leaving an account alone through a genuinely bad stretch can bleed you. The floor is what we call a hard deck.

A hard deck is a spend level you decide in advance and never drop below, regardless of performance. Say you've scaled a campaign to a healthy daily budget and then it goes properly cold. You don't crash it back to near-zero, because at near-zero you can't test your way out and you've effectively switched the account off. Instead you step down to your hard deck, your pre-agreed floor, and you hold there while you go find a new winning creative.

The level is yours to set based on your runway. A newer brand with less cash might hold a modest floor. An established account with reserves might hold a much higher one because it can afford to keep buying data while the creative team works. The point isn't the number. It's that you've decided your floor with a clear head, in advance, so that on a bad day you're following a plan instead of reacting to a feeling.

This is why the boundary matters more than the preference. Whether you scale by 20% or 25%, whether you check at 9am or noon, that's preference, wiggle room, fine. Your hard deck and your no-touch window are boundaries. You don't cross them just because a Tuesday looked bad.

What this looks like over a month

The shape of it, with round numbers: an account opens cold and the early cost per acquisition is rough, well above target. You hold your floor, you don't panic-cut, and you keep feeding the creative pipeline. For a couple of weeks it's unglamorous. Then a new ad lands, takes the spend, and the cost per acquisition snaps down to target. From there it's 20% up, hold, read, 20% up again, and a month that started looking like a write-off finishes spending many times what it did at the start, profitably.

That account didn't recover because someone was clever in the manager. It recovered because nobody panicked and cut it while it was finding its feet.

If you want to pressure-test your own scaling discipline, here's the exercise I'd run: pull your last three budget cuts and ask whether each one followed a real three-day run above target, or just one bad day and a nervous hand. If it's the second more often than not, that habit is quietly costing you, and it's the cheapest thing on this list to fix. It's exactly the discipline we hold the line on when we run accounts, and most of the time the rule is simply: don't just do something, sit there.