Selling into a second country is easy to switch on and hard to verify. The store translates, the currency converts, and orders arrive, so the locale looks done. Underneath, conversion in that locale quietly runs well below the home market, and the reasons are rarely the ones teams guess.
The failures are practical: an address form that fights the local format, a missing payment method that locals consider standard, a tax display that makes the price look wrong. Each one is invisible from headquarters and obvious the moment someone completes a test order as a local.
Complete test orders with real local addresses
Use genuine local addresses, with local postal code formats, province or state equivalents, and the quirks every country has: districts, building names, rural routes. Address validation and autocomplete tuned for the home market often reject, reorder, or mangle foreign formats.
Test the error path too. Enter a slightly wrong postal code and read the message that comes back. If the recovery message names a field that does not exist in the local format, or is written in the home language, a share of local shoppers will simply stop there.
Check that the expected payment methods are present
Every market has payment methods its shoppers treat as non negotiable, and they differ. A checkout offering only cards in a market where a local instant payment method dominates will convert at a fraction of its potential, and the gap looks like weak demand rather than a payment problem.
List the methods a local competitor offers and compare honestly. Then test each method end to end in the local currency, including the redirect out and back, where sessions and carts are most often lost.
Tax and total display must match local custom
In some markets shoppers expect tax included in every price; in others they expect it added at checkout. Showing a local shopper the other convention makes the total feel like a surprise fee, even when the amount is identical, and surprise at the total is a classic abandonment trigger.
The same applies to currency formatting: decimal separators, symbol placement, and rounding. A total formatted in home market style in a foreign currency reads as careless, and carelessness at the payment step costs trust the store can ill afford.
Read the whole flow as a native speaker
Machine translation produces checkout text that is technically correct and subtly wrong: formal where it should be plain, literal where an idiom exists, inconsistent terms for the same step across screens. At checkout, subtle wrongness reads as risk.
Have a native speaker of the locale complete the full flow, including emails and the order status page, and flag anything that sounds off, not just anything incorrect. The goal is a checkout that feels locally built, which is ultimately what the conversion gap is measuring.
Related reading
These earlier guides go deeper on connected questions: The Thank-You Page Is Part of Checkout: Optimizing the Page After the Order, Shopify Markets Meets Checkout: The Cross-Border Settings That Quietly Change Completion, and One Page or Multi-Step? The Checkout Layout Question With No Universal Answer.
Frequently asked questions
Why does a translated checkout still convert poorly?
Usually format and expectation problems: address forms that fight local formats, missing local payment methods, and tax display that does not match local custom.
What should localized checkout testing include?
Complete test orders with real local addresses, every local payment method end to end, deliberate errors to check recovery messages, and a native speaker reading the full flow.
How important are local payment methods?
Very. Some markets treat a specific local method as standard, and its absence looks like weak demand in the data when it is really a payment gap.
Should prices show tax included or added?
It depends on the market's custom and law. The mistake is showing locals the other convention, which makes the total feel like a surprise charge.