When we launched our first global checkout in 2024, we made the classic mistake: we built one checkout and assumed it would work everywhere. It didn't. Conversion rates in Germany were 11 points below the US. In Brazil, they were 19 points lower. The problem wasn't our payment processing β it was our UX assumptions.
The false universality of checkout design
American checkout design conventions are heavily card-centric. Name, card number, expiry, CVC β this is so familiar to US consumers that we treat it as universal. But in Germany, 40% of ecommerce happens via SEPA bank transfer. In Brazil, Pix and boleto bancΓ‘rio are the default. In the Netherlands, iDEAL dominates. Showing a card form first is the equivalent of a Dutch website defaulting to Portuguese.
Research finding: In user testing across 8 markets, participants in Germany and Netherlands reported feeling "mistrusted" when card details were requested before any other option. They associated card-first UI with international (untrustworthy) vendors.
How we rebuilt the checkout
We rebuilt the checkout around three principles:
- Payment method priority by market β the most-used local method is shown first, always. This is determined by a combination of market data and the merchant's own transaction history.
- Address field adaptation β German addresses use different field ordering than UK ones. Japanese addresses are entirely different. We use a configurable address schema per country.
- Trust signals by market β German consumers respond to Trusted Shops badges. UK consumers want to see Visa/Mastercard logos. Brazilian consumers want to see local bank logos. We show what each market expects.
The translation problem nobody talks about
Most internationalisation work focuses on translating strings. We found that button copy has a massive impact on conversion that simple translation misses. "Pay Now" translates to German as "Jetzt bezahlen" β grammatically correct but subtly different in tone. Through A/B testing, we found that "Sicher bezahlen" (Pay securely) outperformed by 14% in Germany and Austria, because security is a higher-priority trust signal in those markets.
Results
The single highest-impact change was simply reordering payment methods by market. No engineering complexity β just data-driven UX. It accounted for 60% of the total conversion improvement and took one engineer two days to ship.
What we're working on next
We're piloting an adaptive checkout that uses real-time signals β device type, time of day, cart value β to dynamically reorder payment methods even within a single market. Early results show an additional 4β6% lift on top of the market-level optimisations.
Ready to optimise your payment flow?
Join thousands of businesses using Zupay to process payments faster, smarter, and at lower cost.