Two ad reports look healthy. Yet their combined purchase count exceeds your store orders. Who is wrong? It is possible that neither has miscounted. The same purchase may simply have more than one claimant.
One customer can appear in both reports
Imagine a simplified, hypothetical day: Google reports 20 purchases, Meta reports 30, and your store records 35 unique paid orders. If both platforms counted only those orders and 15 appeared in both reports, the numbers would reconcile: 20 + 30 − 15 = 35.
In a real report, you cannot automatically label the difference as overlap. First compare the date range, time zone, purchase definition and attribution settings. Also check whether an event was sent twice or a cancelled order still counts as a success.
What decision are you actually trying to make?
On 20 October, IAB opened a draft cross-channel measurement decision guide for public comment. Its central idea is to define the decision first, then choose evidence suited to it. This is not a final standard; comments are open until 4 November.
For a store owner, I would split the issue into three questions. How many orders were paid? Check orders and payments. Which purchases does a platform attribute to its ads? Its report answers that under its own rules. How many purchases would not have happened without advertising? Attribution alone cannot establish that.
Make the comparison consistent before moving budget
Create one store-results table: paid orders, refunds, revenue and contribution after variable costs. Keep platform reports alongside it as separate views. Do not add them to the store total or expect every number to match.
Before a substantial budget increase, write down the additional contribution you expect and how you will test it. A holdout group or geographic test needs sufficient volume and comparable conditions. Comparing one successful week with the previous one does not separate advertising from seasonality, pricing and stock changes.
A useful report makes clear what its data can demonstrate. That keeps one purchase from justifying extra spending in several budget spreadsheets.
Prepared with AI assistance. The example figures are hypothetical; the practical recommendations are the author’s adaptation for everyday store operations.