How to triage a sudden spend spike in a client ad account

First, confirm the number

• Compare the spend in the platform with the billing record for the same day, in the same time zone. A shifted time zone or a day still filling in can look like a spike.
• If the number holds, write it down with the time, the account and the amount. This starts the incident log.

Then find where it went

• Split the spike by campaign, then by ad set, then by ad. Stop at the first level that explains most of the extra money.
• List what changed in the last 48 hours: budgets, bids, new ads, new audiences, and any change made by a person or an AI assistant. Use the change log.
• Read the results next to the spend. Spend up with orders up is a different problem from spend up with orders flat.

Decide, then act

• Check the budget caps. Spend above a cap is a control failure first, and a performance question second.
• Pause only what is clearly wasting money, and log each pause with its reason. Leave the rest running until the cause is clear.
• Every change follows the ad change approval workflow, even under pressure.

Close the loop

• Read the change back the next day, and record the result as measured or not measured.
• Tell the client in writing the same day: what happened, the money involved, what was done and what is still open.