The “can we actually sell there?” test
If a store advertises only to the United States and Europe, a cluster of clicks that appear to originate in those markets but repeatedly shows a different language, currency, time zone, or delivery reality deserves a closer look. The same is true for a software company that cannot onboard or support a particular region.
That does not mean the visitor is fraudulent. People travel. People use workplace VPNs. People shop for gifts. A good report asks whether the signal is consistent with a legitimate buyer before escalating it.
What to compare
- Observed network country versus campaign location and shipping coverage.
- Currency, language, delivery ZIP, and checkout eligibility.
- Product-page depth, add-to-cart behavior, and purchase completion.
- Repeated clicks, session clusters, and whether the same device pattern appears across accounts.
Fraud0’s 2025 research measured invalid traffic across a broad e-commerce and advertising dataset, while Imperva’s report shows how bad bots affect retail and other business logic. Neither study proves that a specific shopper is invalid; they show why stores need traffic-quality controls alongside checkout security.
See the BotRefund workflow or read how to investigate without punishing privacy-conscious buyers.