One thing worth checking on method: many storefronts pick the payment methods, currency and even checkout layout they show from the visitor's IP. Where did the crawler run from for each of the 28 countries? If some checkouts saw a foreign or datacenter IP, the 27% "local method" figure and the 7% showing more than 3 currencies could be off for those markets. It's the same family of effect as Apple Pay only showing in Safari, which you already noted.
Thanks for raising this, Furqan. You’re right: the IP setup matters here. The 27% and 7% reflect what was visible during the crawl, which may differ from what local shoppers see.
I’ll verify the IP location and type used across the 28 countries and clarify this in the methodology. Until then, geolocation remains a potential source of bias in those figures. Thanks for flagging it
Thanks, that's a fair way to handle it. For the check itself, the quickest route is to log the exit IP each crawl worker used and run those through an IP lookup. Country shows whether each market really saw a local IP, and the ASN organization shows whether it came from a cloud provider or a consumer ISP, which is the part storefronts tend to treat differently.
If useful, our API does both on the free plan, and 28 lookups is well inside the daily limit https://ipgeolocation.io/ip-location-api.html. A "what is my IP" check run from each worker would also do the job if you'd rather not sign up for anything.
One thing worth checking on method: many storefronts pick the payment methods, currency and even checkout layout they show from the visitor's IP. Where did the crawler run from for each of the 28 countries? If some checkouts saw a foreign or datacenter IP, the 27% "local method" figure and the 7% showing more than 3 currencies could be off for those markets. It's the same family of effect as Apple Pay only showing in Safari, which you already noted.
Thanks for raising this, Furqan. You’re right: the IP setup matters here. The 27% and 7% reflect what was visible during the crawl, which may differ from what local shoppers see.
I’ll verify the IP location and type used across the 28 countries and clarify this in the methodology. Until then, geolocation remains a potential source of bias in those figures. Thanks for flagging it
Thanks, that's a fair way to handle it. For the check itself, the quickest route is to log the exit IP each crawl worker used and run those through an IP lookup. Country shows whether each market really saw a local IP, and the ASN organization shows whether it came from a cloud provider or a consumer ISP, which is the part storefronts tend to treat differently.
If useful, our API does both on the free plan, and 28 lookups is well inside the daily limit https://ipgeolocation.io/ip-location-api.html. A "what is my IP" check run from each worker would also do the job if you'd rather not sign up for anything.