Sales Dropped and Spend Didn't: The Order to Check Things In

It always looks the same. Spend is unchanged, the campaigns are unchanged, nobody touched anything, and orders are down. The first reaction is to open Ads Manager and start shuffling ad sets.

That is the worst place to start. The account can be the cause, but it is less likely than the three layers before it, and it is the one place where a wrong change costs you immediately. The order below is the same one I use in my Pre-Account Audit. If orders went all the way to zero rather than dropping, start with the five checks for ads spending with no orders instead.

Why did my sales drop when my ad spend stayed the same?

In most cases because something changed outside the ad account: tracking, delivery, stock, an offer, a price. Compare the last 14 days with the 14 days before them and walk the four layers in order: measurement, back end, offer and page, then the account.

Did sales really drop, or did tracking break?

Check that before anything else. In Egypt the gap between the orders Meta reports and the orders the store records regularly reaches 40%, so a flat business can look like a falling one. The three things that most often make steady sales look like a drop:

• The pixel broke or was changed in the theme after an update, so orders happen and aren't recorded.
• UTMs broke, so ad orders now show as direct in Shopify.
• The attribution window changed, or the selected period has a different number of days.

The check is quicker than the explanation. Open the sales report from the store itself, not from Ads Manager, and compare total orders. If the store number is flat and the Meta number is down, there is no sales problem. You have a measurement problem, and the fix is completely different. I covered how far reported numbers drift from cash in what ROAS really means on cash on delivery; on one account a reported 5.48 ROAS turned out to be a real 1.80 once view-through and cancelled orders were removed.

Could the drop be coming from delivery, stock or confirmation?

If sales really did drop, this layer is where the answer most often comes from, and it is the layer people look at least. A confirmation call that slips from 2 hours to 24 hours is enough to cancel a meaningful share of orders.

Delivery rate

Orders can hold steady while the cash coming in falls, because a bigger share of them comes back. That never shows in Ads Manager; the account keeps saying everything is fine.

Pull the delivery rate for the last two weeks and compare it with the two weeks before, split by region and by courier. If a courier changed the driver in one area, it shows up as a sales drop when demand hasn't dropped at all.

Stock

If the best seller sold out, or half its sizes are unavailable, the ad keeps running, visits keep coming and orders fall. It is common enough that it should be checked before anything in the account.

Confirmation

When confirmation slows from 2 hours to 24 hours, more orders get cancelled. That is an operations change, not an ads change, and its effect appears about 7 days after it happened.

What changed in what the visitor sees?

This layer has one question: what is different on the screen the visitor lands on? An offer that ends hits within 1 day; a price increase shows over 3 to 7 days.

How do I tell whether the problem is the ads or the website?

Read traffic and conversion rate together over the same 14 days. If traffic holds and conversion falls, the problem is the page, the offer or the price. If traffic falls and conversion holds, the problem is in the account.

That split alone saves days. Most of the time lost on a drop goes into changing something in the account when the problem was on the product page.

How do I know if my creative is fatigued?

Only now do you open the account, and the first thing to look at is the creative, not the ad sets. Fatigue is two numbers moving together: frequency rising and click-through rate falling while CPM stays roughly flat. The same people are seeing the same ad one more time and responding less. Re-cutting the first 3 seconds of the same video buys about 7 days, and then the same problem is back.

So the fix isn't a new caption or thumbnail. It is a new angle. That is why I plan 40 to 60 new ads a month for an account, each with a written test, rather than waiting for the winner to die.

Audience saturation

Reach plateaus and cost climbs. The account has used up the room in the current audience and needs a wider one or another platform, not more budget in the same place.

A change you forgot

Open the change history in Ads Manager and set the range to start a week before the drop. An ad set paused, a creative turned off, targeting edited, budget moved. Most of the cases where "nobody touched anything" turn up a change logged with the date and the time.

What should I do once I find the cause?

Write it down. Every rise and every fall gets a reason in the same sheet: the month, the spend, the sales, and one line with the cause. Without that column, 60 days later someone looks at the same drop and calls it a dead month, when it was an offer that stopped or a product that sold out.

The difference between those two explanations is the difference between a right decision and a repeated one. And when sales recover, I raise the budget in steps of 20 to 30 percent, only while profit per delivered order holds.

If you want a second pair of eyes on a drop, send me the store link through the free audit. I reply within 24 hours, and within 48 hours you get three numbers: your maximum cost per order, your ROAS after delivery, and your biggest leak.