Your dashboard says payouts are paused, last week's orders are sitting somewhere you can't reach, and the email you got names no date. The ads are still running at yesterday's rate, the fulfillment bill is still due this week, and support has sent you the same paragraph twice. This is the order to work in for the next 14 days, sequenced by what is draining cash fastest rather than by what feels most urgent.
3 different events get called a freeze, and they cost different amounts
Work out which one you're actually in before you touch anything, because the correct first move is different for each. The notices are written to be vague, so the state you're in is usually clearer from the dashboard than from the email.
Read it yourself instead of waiting for support to explain. Check whether a live test order still completes at checkout, whether the payouts page still shows a next payout date, and whether the account status reads restricted, under review, or closed. Those 3 readings put you in one of the states below, and they are what you quote back in your first email instead of asking what happened.
- Payouts paused, charges still working. Orders keep landing in your balance and nothing moves to your bank. Revenue is intact on paper and your cash is not. This is the most common version and the most survivable.
- Charges disabled, existing balance still payable. Checkout fails for new customers. Now the storefront is the emergency and the bank account is the smaller problem, because the hole gets bigger every hour the buy button is dead.
- Account closed, funds held. The notice says the business can no longer be supported, cites a clause or a restricted category, and holds the balance against refunds and disputes that arrive later. The hold length that applies to you is the one written in your notice, not the one somebody quoted in a forum thread 2 years ago.
Hour 1: cut the spend, not the campaigns
The freeze doesn't touch your ad account, and that's exactly the problem. Meta keeps buying at yesterday's pace against a bank balance that stopped refilling this morning, and nothing in the ads manager knows anything has changed.
Switching everything off is the reflex and it's usually the expensive choice. You lose the learning, you lose the revenue that would have covered the gap, and you restart into worse delivery a week later. Cut to the ad sets that return cash fastest, pause the ones you were still testing, and leave the proven ones alive at a reduced budget.
Then put a hard ceiling above all of it. An account level spend limit is the only cap that holds regardless of what happens inside the campaigns, and it's the one control an auto rule or a stray budget edit can't step over while your attention is on the processor. The mechanics are in the walkthrough of how account spend limits work and how to raise them again, and during a cash gap it's worth setting deliberately low and moving it up by hand.
There's a second failure most operators meet only once. Ad platforms bill on their own schedule against the card on the account. If the frozen balance was feeding the account that card draws from, the charge fails, and a failed payment is a restriction risk in its own right. Move a card the freeze can't reach onto the ad account today, before the next billing threshold hits.
Hour 6: send the email that produces a date instead of a template
Reviewers work through queues and answer specific questions. A long message about how unfair this is buys nothing except a slower position. Send 1 short email with attachments, and make it easy to say yes to.
Send it in 1 thread with the documents attached, not behind a link the reviewer has to log into and not spread across 5 messages. Then stop sending. Every follow up buries the attachments under your own words and gives whoever finally opens the thread more to read before they reach the evidence. This is what goes in it.
- Fulfillment proof for the last 30 days: orders shipped, tracking numbers, carrier name.
- Your unshipped backlog stated honestly, with the date it ships and why it slipped.
- Supplier invoices or a purchase order showing you can fulfill what you've already sold.
- The refund and cancellation policy URL, plus your current refund rate and dispute rate as your own dashboard reports them. Understating either is the fastest way to lose the reviewer.
- A named human, a phone number and a support address, all matching what's on the site and on the statement descriptor customers see on their card.
Ask for 3 things in writing, and don't argue policy
Close the email by asking for the specific reason for the action, the exact date the review or hold ends, and whether charges can be re-enabled while the review continues. Those 3 answers are what your next 2 weeks depend on, and a verbal answer from a chat agent is worth nothing when you're planning cash against it.
Don't dispute the policy itself. The person reading your message can't reverse a category decision and isn't the audience for that argument. Give them evidence that the business is real, that customers get what they paid for, and that money going out later is unlikely to exceed the balance being held. That's the entire question a risk team is answering.
Day 1 to 3: do the arithmetic before you make any other decision
The question isn't whether you eventually get the money. It's how many days you can operate without it. Take the cash you hold outside the processor, subtract what you owe in the next 14 days, and divide by your real daily burn: ad spend plus cost of goods plus shipping plus anything on a standing order.
Suppose the freeze is holding $40,000, you're spending $2,000 a day on ads, and you have $18,000 in the operating account. The ad spend alone eats that in 9 days, before a single supplier invoice clears. That is your actual deadline, and it usually falls well before the date in the notice. Everything else in this article is about moving that deadline further out.
There's a trap on the other side of the ledger. Refunds and disputes keep debiting while payouts are paused, and if the held balance can't cover them the processor goes to your bank account for the difference. Stopping refunds to protect cash is the single worst move available: unhappy customers who can't get a refund file disputes instead, and dispute rate is a number the review is judging. Keep refunding, and keep refunding fast.
Day 3 to 7: stand up the second rail
A backup processor you opened once and never used isn't a backup. Dormant accounts get closed for inactivity, and an underwriter looking at zero recent volume treats you as a new applicant on the day you're least able to wait. The version that works is a live secondary carrying a small share of daily volume all year, so the switch is a routing change rather than an application.
- Keep the secondary processing real orders continuously, not sitting at zero. Route a slice of traffic to it every week so it has recent history when you need to lean on it.
- Saved cards don't travel by default. Stripe supports a PCI compliant data migration for moving stored card data to another provider, and it runs as a written request between the 2 providers rather than a button in the dashboard. Start it early, and don't assume it stays available once an account is fully closed.
- Recurring billing doesn't move with the cards. Subscription schedules, trial end dates and next charge dates live inside the processor you're leaving, so a card migration hands you card records and nothing that knows when to charge them. If you sell a subscription offer, rebuilding those schedules is the long job, and every day it takes is a day of rebills that quietly don't happen.
- If the answer turns out to be a second storefront on a different domain, the ad side breaks before the payments side does. A new domain means verification and pixel work again, and the piece on why domain verification and pixel setup decide where your ads can point is the checklist to run before you move a single campaign.
- Match the statement descriptor to the brand name customers see at checkout on both rails. Descriptor confusion is a dispute generator, and disputes are the metric that decides how the current review ends.
The line between redundancy and evasion
Redundancy built before an enforcement event is ordinary operations. A second processor, a second entity with its own real trading history, a standby storefront: businesses build those because outages happen, and a rail with its own volume, its own descriptor and its own paperwork is not what an underwriter is looking at when they look for evasion.
What you don't do is open a fresh account for the same business, on the same bank account, the same device and the same domain, after being closed. Processors link accounts on exactly those signals, and closing both is the standard outcome rather than the unlucky one. The same line exists on the ad side, where provisioning spare assets in advance is normal and rebuilding a lookalike account on the same card and pixel after a ban is circumvention. The rundown of what a legitimate asset swap looks like mid campaign draws that distinction on the Meta side, and it's the same distinction here.
Day 7 to 14: what the freeze exposed about your ad structure
A processor freeze stops being a payments problem before the first day is out. After that it's a media buying problem, because every part of your setup that depends on cash sitting somewhere you don't control has just turned into a second balance you can't reach. That's worth auditing now, while the memory is expensive and fresh.
Start with a question most operators never ask their ad account provider: where does the ad money physically sit, and who can move it. If you pre-fund a balance held by somebody else, that's cash on a shelf you can't take back on the exact day you need it. If you pay a percentage of ad spend, the bill grows the moment you scale to trade your way out of a hole, which is the worst possible time for a variable cost to expand. The numbers on that second one are laid out in the breakdown of what a percentage cut of ad spend actually costs at real spend levels.
The honest trade off runs the other way at low spend. Take $299 a month against a 5% cut, which sits inside the 0.5% to 20% range providers in this market quote. The crossover is around $6,000 of monthly ad spend. Below that, a percentage is simply cheaper and you should pay it. Above it, a flat monthly rate stops growing with your budget, and on the top tier your own card goes on the account so Meta bills you directly rather than you loading a balance somebody else holds. During a cash freeze the second half of that matters more than the first, and you can see how it's split across the account tiers and what each one hands over.
None of this makes a freeze go away. It changes how many days you can absorb one, which is the only variable you actually control.