A merchant’s guide to Spain and the United States
Spain vs. the U.S.:
What changes at checkout
A new market can make a familiar checkout behave very differently. A Spanish customer may leave your site to approve a payment in a banking app. An American customer may complete the same purchase without that step. Either order can still become a loss if the wrong person pays—or the right customer never receives what was promised.
For merchants expanding across the Atlantic, the challenge is to protect the sale without obstructing it. That means matching fraud controls to the payment, understanding who bears a loss, and judging success by the money the business keeps.
Spain & United States · Merchant payments and fraud · Updated August 31, 2026
Where the markets diverge
Most controls travel. A few assumptions should not.
A checkout can be translated overnight. The assumptions behind it take more work. A merchant used to U.S. address checks may reject a Spanish buyer simply because the issuer returns less information. A merchant accustomed to European authentication may add a challenge to American orders that do not need it. Both mistakes turn a fraud rule into a lost sale. [2], [4], [5]
The useful comparison is the same payment activity in each market. The table below isolates differences that can change approval rates, implementation work or cost. It does not rank countries by safety: a merchant’s products, customer mix and delivery model still determine much of the exposure.
On small screens, scroll the tables sideways to read both markets.
Loading..........
The Data is Not Available
Authentication: sources 1–4. AVS: 5–6. Pricing and routing: 10, 21. [1], [2], [3], [4], [5], [6], [10], [21]
The shipping address does not settle the authentication question.
One control framework, adapted locally
Match the check to the risk it can actually see.
The core fraud controls are largely shared. What changes is their availability, when a provider requires them and how much friction customers encounter. The expensive mistake is to treat any single result—an address match, a token or a bank approval—as an answer to every question about an order.
Loading..........
The Data is Not Available
Terminal controls: sources 8–9. Online screening: 3, 6–7. Tokens and renewals: 11–13. [8], [9], [3], [6], [7], [11], [12], [13]
Successful 3D Secure can shift liability for qualifying fraud disputes, subject to the network rules and transaction outcome. It does not protect a merchant against every dispute reason. A buyer can approve a purchase and still receive the wrong goods, wait beyond the promised delivery date or question a charge made after cancellation. Those problems need fulfillment and billing evidence, not another authentication signal. [3]
For subscriptions, the first purchase is also the setup for the next one. Capture the customer’s permission and complete applicable authentication while they are present. Network tokens can help maintain usable credentials, but a renewal that requires fresh authentication needs a clear route back to checkout. Separate those recoverable failures from cancellations so that revenue recovery does not become unwanted billing. [11], [12], [13]
Make the configuration prove itself
What should change before your first sale?
A launch test is more revealing than a feature list. Open the scenarios below and use them to check the weak points in the actual payment flow—not just whether the happy path produces an approval.
Run an online-card purchase that requires a 3D Secure challenge. Leave for the banking app, return, cancel once and retry. The amount, basket and order reference should survive; retries should not create duplicate orders. This is especially consequential for Spain’s SCA flows, and still matters whenever a U.S. payment is challenged. [2], [3]
Test AVS match, mismatch and unavailable responses separately. Then review them alongside customer, device and delivery history. Broad U.S. coverage makes AVS useful; different issuer coverage means an unavailable result cannot sensibly be treated as a universal decline rule. Inspect repeated low-value attempts separately for card testing. [5], [6], [7]
Test an expired credential, a renewal that requires authentication and a canceled subscription. The first two may call for an update or customer action; the canceled subscription should not be retried. Confirm that the gateway sends the correct stored-credential indicators and that support can see what the customer originally agreed to. [11], [12], [13]
Run a pending bank debit, a returned collection and a refund. Confirm which status permits shipment, who receives the alert and how the order is reconciled. At a physical checkout, also rehearse failed terminal reads so staff do not turn a declined chip transaction into improvised manual entry. [8], [16], [18]
Bring those scenarios into your account review. When you apply for a merchant account with Durango Merchant Services, include the markets, products, delivery times and billing model you plan to support. That makes the discussion about the business you will operate, rather than volume alone.
Bank payments need their own comparison
A low-cost collection is not a guaranteed collection.
“Pay by bank” can mean that the customer sends money or that the merchant collects it. The distinction matters most when goods are difficult to recover. A quick account check can be useful without making the proceeds final; a low collection fee can be attractive without compensating for a large returned order.
Compare consumer bank debit with consumer bank debit.
Loading..........
The Data is Not Available
The same operational discipline belongs behind either debit: monitor returns, assign responsibility for investigating them and decide how much unrecoverable value to release before the risk is acceptable. For a high-value first order, stronger identity or ownership evidence and staged fulfillment may be worth more than a marginal fee saving. That is a business decision about exposure, not a claim that one scheme is universally safer.
Transfers solve a different problem.
For euro transfers in Spain, Verification of Payee can warn when the recipient name and account details do not match. For U.S. ACH, Nacha’s June 2026 expansion extends risk-based fraud monitoring to the remaining covered non-consumer originators and service providers. These are different mechanisms, not equivalent guarantees against a fraudulent invoice. Independently verify changed supplier bank details through a trusted contact route. For incoming payments, reconcile against bank or provider confirmation rather than a customer’s screenshot. [19], [20]
Where Bizum fits
Bizum may suit one-off purchases from Spanish customers, but check the provider’s implementation before treating it as a replacement for cards or recurring debits. Stripe and Adyen document Bizum integrations without recurring billing or separate manual capture. Entity and category eligibility also matter. Stripe documents transaction claims, so the absence of card-style chargebacks should not be read as the absence of disputes. [14], [15]
Two ways to see the commercial trade-off
What is a better checkout worth?
Processing fees are easy to compare. Lost legitimate orders are less visible, and fraud losses may appear weeks later. The illustrations below put those costs on the same footing: contribution after merchandise cost, processing, unrecovered fraud and the entered dispute costs. They are hypothetical merchant scenarios, not Spanish or U.S. market averages.
One change at a time: extra contribution per 1,000 checkouts
USD · Each bar changes one assumption while holding the others fixed.
A 1-point completion gain means 10 more completed orders here. The interventions are different sizes; this shows their dollar effects, not which is easiest or cheapest to achieve.
How much improvement pays for a higher fee?
USD gain or loss versus Setup A · Horizontal axis: Setup B checkout completion rate.
Setup B charges more but assumes a lower fraud-loss rate. At the same 62% completion rate, it keeps $148.80 less. It breaks even at about 62.51% completion; at 65%, it keeps $730.50 more per 1,000 sessions. A higher-priced setup therefore needs a measurable commercial benefit, not just a reassuring feature list.
The assumptions behind both charts
1,000 eligible checkout sessions; $80 average order; 40% merchandise margin. Setup A: 62% completion, 2.4% + $0.25 processing, 0.35% unrecovered fraud loss and $60 other dispute costs. Setup B: 2.8% + $0.25 processing, 0.25% loss and the same $60 costs. Percentage fees and losses apply to captured sales. These fixed illustrations do not change when you edit the separate calculator below.
Test your own economics
What would have to improve for your business?
Compare two proposed setups for the same market and the same 1,000 eligible checkout sessions. Use your own quote and observed performance where possible. Keep products, customer mix and merchandise margin comparable; a change in the audience can otherwise look like an improvement in the payment setup.
Compare two setups for the same 1,000 eligible checkout sessions.
Setup A and Setup B are not country benchmarks. Replace the examples with your own assumptions. Nothing you enter is sent or saved by this calculator.
Live results require JavaScript. If they do not appear, use the formula below or ask your site administrator to check the HTML widget’s script permissions.
Estimate based only on the inputs above. It is not a forecast or a provider recommendation.
Setup A
Estimated contribution retained / 1,000 checkouts- Captured orders
- Captured sales
- Contribution before payment costs
- Less processing fees
- Less unrecovered fraud principal
- Less other payment costs
Setup B
Estimated contribution retained / 1,000 checkouts- Captured orders
- Captured sales
- Contribution before payment costs
- Less processing fees
- Less unrecovered fraud principal
- Less other payment costs
Setup A: where the contribution goes
Setup B: where the contribution goes
Both charts use the same scale. Navy: contribution before payment costs. Orange: deductions. Teal: contribution retained. A negative result appears in red. All values also appear in the breakdown above.
How the estimate works—and what it leaves out
Captured sales = 1,000 × completion rate × average order. Starting contribution = captured sales × margin. Subtract percentage and fixed processing fees, unrecovered fraud principal, and the entered total for other payment costs.
The margin input must already reflect goods, fulfillment and ordinary returns, but must not already deduct the fraud principal entered separately. Count each cost once. This estimate excludes taxes, fixed overhead and financing costs unless you incorporate them appropriately. Reserves affect available cash and need separate review.
Compare similar customer groups over the same observation period, allowing time for refunds and disputes. Holding basket size and margin constant makes the two setups easier to compare; actual payment changes may also affect the customer mix.
Before acting on a result, follow the resulting orders long enough to see returns and disputes. Report bank-authentication failures, issuer declines and your own fraud-rule blocks separately. That is how you find out whether a new configuration recovered good orders, merely moved a failure elsewhere or accepted losses you have not seen yet.
Keep the account healthy
The payment is only the beginning of the evidence.
Your fraud system and customer-service team should be able to tell the same story about an order. Retain the agreed terms, promised delivery date, tracking or relevant service-use evidence, customer correspondence and cancellation history. Those records help distinguish stolen credentials from a delivery complaint or an unwanted renewal; they do not guarantee a dispute win. [3]
This distinction also matters to account monitoring. Visa’s VAMP framework combines specified fraud reports and disputes for covered card-not-present transactions. A low fraud-loss figure alone is therefore an incomplete view of account health. Ask the acquirer how it counts disputes and when it intervenes, then watch the underlying causes rather than waiting for a network threshold to become the first warning. [22]
Plan the payment side of your expansion
Build the account around the business you intend to run.
A retailer shipping immediately, a travel company collecting deposits and a subscription business carry different exposure in either country. Give the account review that context: what you sell, where your customers are, when you deliver and how you handle refunds.
Apply with Durango Merchant Services to start the conversation about the payment requirements of your expansion.
Account approval, available services and terms are subject to underwriting and provider eligibility.
Evidence & scope
Sources behind the analysis.
Research checked August 31, 2026. Tables compare equivalent payment activities; they do not rank national fraud performance. Charts are illustrative calculations, not market statistics. Provider documentation describes specific implementations, not capabilities available from every processor. Recommendations are our analysis; confirm applicable rules and contractual responsibilities with your provider.