Your Purchase Numbers Are About to Move. Do Not Panic
Table of content:
A platform ships a reporting change in the middle of November, your purchase count moves, and you have eleven days to decide whether something broke or whether the number is simply being counted differently now.
Founders lose real money to this every year, because the instinct when a number drops is to change something, and changing something during peak costs more than the drop did. This post covers how to tell the two apart and what to settle before the weekend.
Why do platforms change things right before peak?
Because their release calendar is not built around your trading calendar, and because peak traffic is when a change gets tested at scale.
It is reasonable to find this frustrating. It is not reasonable to plan as though it will not happen. Something shifts in measurement most Novembers, and the brands that handle it well are the ones who decided in advance how they would respond.
Operators have been reporting two changes this month: an adjustment to how some conversions are reported, and budget notifications behaving differently where daily caps are concerned. Treat the specifics as worth verifying in your own account rather than taking anyone's summary at face value, this one included.
How do you tell a reporting shift from a real drop?
By checking whether the money moved.
Your platform reports attributed conversions. Your commerce platform reports orders and revenue that actually happened. When the first moves and the second does not, you are looking at a counting change. When both move together, something real happened.
Three checks settle it quickly. Compare platform reported purchases against your store's order count for the same window. Check whether the change appears across every campaign at once, since a genuine performance problem rarely lands everywhere simultaneously. And check whether it coincides with a date rather than with anything you did.
A shift that starts on a specific date, appears account wide and leaves your order count untouched is reporting, not performance.
What should you not react to?
Anything that moves your reported numbers without moving your bank balance.
The expensive responses are predictable: cutting budget on a campaign that is actually fine, restructuring to chase a number that will revert, switching attribution settings mid season, and turning off a campaign that was carrying the account. Each one costs learning during the weeks when learning is most expensive to rebuild.
The rule worth writing down is that no structural change gets made in response to a reporting movement until the order count confirms it. That single sentence prevents most of the damage.
Why would spend exceed the cap you set?
Daily budgets are generally a target that the system averages over time rather than a hard ceiling for any given day, so a high intent day can overspend and a quiet day can underspend.
That behaviour is ordinary and it matters more during peak, because the days when the system wants to spend ahead are exactly the days when volume is highest. A founder watching a daily figure can see a number well above what they set and conclude something is broken.
If total spend across the period is what you care about, use the budget type that controls the total rather than the daily rate, set it before the weekend, and watch the period figure rather than the daily one. The pacing side is covered in Q4 budget pacing.
How do you keep a baseline that survives the change?
Record the numbers now, with the settings beside them.
Before the weekend, write down your current attribution setting, your platform reported purchases and revenue, your store's orders and revenue, and the gap between the two as a percentage. That gap is the useful artefact. If it widens after a platform change, you know the change affected you and by roughly how much.
Then leave the attribution setting alone until January. Changing it mid season makes November incomparable with October, which costs you the ability to say afterwards what actually worked.
What is worth acting on?
Three things, all confirmed by your own order data rather than by the platform.
Orders falling while spend holds. Cost per acquisition passing the ceiling you set, measured on real orders. And delivery stopping, meaning impressions collapsing at an unchanged budget, which usually points at a policy or payment problem rather than at performance.
Everything else can wait until January, and most of it will have resolved itself by then.
What to agree before the weekend
Who decides, on what evidence, and what nobody is allowed to touch.
Name the person who can change budgets during Cyber 5. Agree that structural changes need order level confirmation. Write down the two or three numbers that would trigger action and the thresholds that go with them. The wider version of this, including the decision rhythm for the five days, is in peak escalation rules and the Cyber 5 day by day plan.
The difference between attributed and incremental revenue, which is the deeper version of this question, is covered in incremental attribution. The full measurement framework sits in the Q4 Profit Playbook, free and written for founder led DTC brands on Shopify at $5M to $30M.
Frequently asked questions
My reported purchases dropped. How do I know if it is real?
Compare platform reported purchases with your store's order count for the same window. If orders held steady while the platform number moved, you are looking at a counting change.
Should I change attribution settings during Black Friday?
No. Changing them mid season breaks comparability with your own baseline, so you lose the ability to say afterwards what worked. Record the current setting and revisit in January.
Why is my daily spend higher than the budget I set?
Daily budgets usually function as a target averaged over time rather than a hard daily ceiling, so high intent days can overspend. Use a total budget type if the period figure is what matters to you.
What should trigger a change during peak?
Orders falling while spend holds, cost per acquisition passing your ceiling on real order data, or impressions collapsing at an unchanged budget. Reported metrics alone are not enough.
How do I protect my reporting from a mid season platform change?
Record platform numbers, store numbers and the gap between them before the weekend. If the gap widens later, you can size the effect rather than guess at it.
Get weekly expert insights!
Built from scaling real brands
Turn your ad spend into real growth.
At Webtopia, we don’t just run ads. We build scalable growth systems designed for ambitious DTC brands. By combining performance marketing, creative strategy, and data-backed execution, we help founders scale without sacrificing profitability. Our clients see an average 6X blended ROAS every month, because great brands deserve more than short-term wins.
Book your call today and let’s build your next growth chapter together.