Black Friday Is the Only Time We Turn Off Cost Controls: The Peak-Week Pacing Playbook

The first thing I look at when I open a new ad account is the bidding. Not the creative, not the audiences. The bidding. It tells me almost everything about how the previous manager thought about risk.

And nine times out of ten, on an account that's been run by someone careful, I find cost controls running on every campaign, all year, no exceptions. Cost caps, bid caps, minimum ROAS. Discipline everywhere. Which is mostly the right instinct.

But then I scroll back to the previous November, and I find the exact same cost caps still bolted on through Black Friday weekend. And that's the moment I know the account left real money on the table, because Black Friday is the one window of the year where those controls are working against you.

Here's the take I'll defend to any founder: for 363 days of the year your job is downside protection. For about two days a year, your job is the opposite. The risk flips. And running the same bidding through both is how careful operators quietly underperform when it matters most.

Why the risk flips for two days

Think about what a cost cap actually does. It tells Meta: do not bid for this customer if it costs more than X. Every other day of the year that's exactly what you want, because the danger is overspending into customers who weren't worth it.

On Black Friday and Cyber Monday, the danger is the reverse. Consumers have their cards out. They've been waiting weeks for your offer. The buying behaviour sits completely outside the normal 363-day pattern. The thing that can hurt you most isn't a bad CPA, it's an hour where demand was there and your bidding refused to chase it.

Let me put that in money. Say you'd normally do ~$8k of revenue in a peak hour on Black Friday. If a cost cap holds your delivery back for three of those hours because the auction got expensive, you didn't save money. You quietly skipped ~$24k of revenue that was sitting right there. A bad CPA day costs you margin on the sales you did make. A missed peak hour costs you the whole sale, and you never see it in the account because it simply never happened.

That asymmetry is the entire argument. A missed BFCM hour costs more than a bad CPA day. Once you believe that, the playbook writes itself.

The build: one simple campaign, plus a phantom

I don't do anything clever for peak week. Clever is fragile. Simple is what survives a high-pressure weekend.

The core is one consolidated campaign with as much of your creative loaded into it as the structure sensibly allows. You only split a campaign or ad set out when you genuinely need a different bid, and the usual reasons are a different margin story. So a separate campaign for lapsed or retention customers, and a separate one for international if your delivery costs there are different. Beyond that, resist the urge to over-segment. Fragmented budgets behave worse under pressure, not better.

Run that core campaign on your cost control as usual, right up to the weekend.

Then build the part most people skip. Duplicate that exact campaign, set the copy to highest-volume bidding, and push it through Meta's review and approval days ahead, then leave it switched off. A phantom campaign sitting in the wings, already approved, doing nothing.

The reason is timing. If your cost-capped campaign starts choking on delivery mid-Friday because demand has blown past your cap, you do not want to be scrambling to launch something new, then waiting on review while the best hours of the year tick by. You flip the phantom on. It's already cleared. It picks up the volume your capped campaign won't chase. That single move is the difference between catching peak demand and watching it pass.

"Surfing" spend, and the 1% rule

Most of the year I'm a firm believer in boring budget moves: 20% up on a campaign that's hitting its target, 20% down on one that isn't, judged against a clear cost-per-acquisition North Star. No doubling, no theatrics. I've watched too many people try to "catch a falling knife" by yanking budgets around daily and make everything worse. On a normal day, the account corrects itself faster than you can.

Peak week is the one exception where I'll surf. Surfing means doubling a budget two or three times in a single day, the kind of move that's reckless on an ordinary Tuesday. The reason it's safe here and nowhere else is the time pressure: you've got a tiny window to spend into enormous demand, and the slow 20% method simply can't move fast enough to keep up.

That's it. That's the whole list of times I'll surf: Black Friday weekend, and the occasional one-off two or three day event. Add it up and you're surfing maybe 1% of the year. The other 99%, the controls stay on and the moves stay small. Anyone telling you to surf budgets as a general scaling tactic is selling you volatility dressed up as aggression.

The Tuesday-morning pulldown

Here's where most of the damage actually happens, and it's got nothing to do with how aggressively you scaled. It's the comedown.

The single biggest mistake I see is leaving the Black Friday weekend budget running into Tuesday morning. The cards have gone back in wallets. Demand has snapped back to normal. But the budget is still sized for peak, and the bidding is still loosened off, so you spend Tuesday torching money at a CPA that made sense Saturday and makes no sense at all now.

So I treat the wind-down as deliberately as the ramp. Here's the pacing I'd run:

• The night before, not the morning of. Set every daily budget the evening before each peak day, so it's live when the day starts. And check the time zone your ad account is set to, because if it's on a different zone to you, your "Friday budget" can kick in hours off and quietly ruin your pacing. This one catches people every single year.
• Friday end of day: pull back and tighten. As the Friday rush eases, ease budgets down a notch rather than letting them run flat into a quieter overnight.
• Sunday: open back up into Cyber Monday. Loosen again as Monday demand builds, repeating the same ramp you ran for Friday.
• Monday night: this is the one that counts. Pull your budgets back down to normal and switch the phantom highest-volume campaign back off before you go to bed. Do not leave it for the morning.
• Tuesday: back to 363-day discipline. Cost controls on, budgets at their everyday level, the boring 20% moves resumed. The party's over and your bidding should know it.

None of this is exotic. It's just treating the two days where the risk inverts as genuinely different from the other 363, instead of running one setting all year and hoping it flexes.

If you want a sanity check on your own setup, here's a simple exercise: open your account and look at last November's bidding, hour by hour across that weekend. Were your controls helping you catch demand, or quietly capping it? Then look at the Tuesday after. If your spend was still sized for peak while your sales had already normalised, you've found exactly where next year's money is hiding. That's the kind of thing we map out for clients running Meta through peak season, but it's an afternoon's work to check it yourself, and well worth doing before the calendar turns.