PAYMENT QA GUIDE

Choosing Payment Test Cards: Local QA vs Provider Sandboxes

A “test card” is not one category of data. Local format fixtures, provider sandboxes, and network certification environments answer different questions, and mixing them creates misleading results.

Start with three layers of test data

Layer one is local format data for length, Luhn, formatting, and masking. Layer two is provider-owned sandbox data for simulated success, decline, 3DS, and risk outcomes. Layer three is acquiring or network certification, which usually requires dedicated accounts and test plans.

TestPAN Studio handles layer one and links to official documentation for layer two. Custom-prefix output does not represent an account and must not be used to predict a payment outcome.

What an end-to-end matrix should cover

Beyond a successful payment, cover generic decline, insufficient funds, expired details, verification failure, 3DS challenge, duplicate submission, timeout, and idempotent retry. Exact values and companion fields must come from the current PSP documentation.

Every fixed scenario in automation should record its provider, expected result, and source so a value for one gateway is not accidentally reused with another.

Test data still deserves safe handling

Mask sequences by default and keep only the necessary last four digits plus a correlation ID in logs. Copying or exporting a complete value should require a deliberate user action.

Never collect real cardholder data to test a validator. API input should remain string typed, body and batch sizes should be capped, and responses should use no-store caching.

Keep official scenarios traceable

Stripe, Adyen, and Braintree maintain separate test-card documentation. Link the source from your test suite and review cases when rules change instead of preserving an unversioned copy forever.

The fixed scenario library here is for format regression and fast navigation. Always get CVV, expiry, regional constraints, and account configuration from the provider page.