A CVV test in UX is a check of how the card security code field behaves in a payment form: the label, the input type, the error message, and the result when you submit known test card numbers in a sandbox. The aim is a field a shopper fills once, without a guess, and a system that never stores the code after authorization.
It collects three digits for most cards and four for American Express. It sits next to the card number and expiry, opens a numeric keypad on mobile, and accepts a paste. Nothing else.
inputmode="numeric", autocomplete="cc-csc", and a length limit that follows the detected card brand.Run every test in a sandbox with published test card numbers. Stripe, Adyen, and other processors publish numbers that trigger approvals, declines, and specific error codes, and none of them need a real security code.
Test values are not real codes. Most sandboxes accept any three digits on a test card, which is why you should also push invalid strings through the field to see how it handles them.
Three causes cover most cases: the card number is not a sandbox number, the account is in live mode, or the processor checks the code against a rule your test card does not match. Check the mode first, then the card number.
Most failed entries come from the form, not the shopper. Fix these before you run another A/B test.
Good error copy names the cause and the fix: "Check the 4-digit code on the front of your Amex card." Keep it under one line on a phone.
PCI DSS Requirement 3.2 bans storing the CVV after authorization, in any form. That rule shapes the interface: no "save this code" checkbox, no code in your logs, no code in analytics events, no code pasted into a support ticket.
Strip the value from session recordings. If your analytics tool captures form input by default, exclude the CVV element before you ship, and check the exclusion again after any tag change.
Never type a real card's security code into a test environment. Test values exist for that job, and a real code in a sandbox is a data handling problem, not a test.
Track four numbers: error rate on the field, time spent on the field, abandonment at the payment step, and the split between "wrong length" and "wrong code" errors. A field with a high error rate points to a label or length problem, not a shopper problem.
Test one change at a time. Switching the label, the keyboard, and the error copy in the same release leaves you unable to tell which change moved the number.
One box. Three separate boxes add tab stops, break paste, and confuse autofill.
Helper text under the label works better than a link, which pulls shoppers out of checkout. Use an icon with a tooltip only when space is tight.
Yes, if the autocomplete value stays correct. Test with a screen reader and with paste before you commit, since a mask can hide typing errors from the shopper.
No. It confirms the code matches the issuer record at that moment. It does not prove the shopper owns the card, so pair it with address verification and 3-D Secure.
You will receive the digital voucher immediately.
All your data is transmitted with unbreakable SSL-RSA encryption.
You can give away, resell or keep the vouchers.
We are available for you 24 hours a day.
Be informed as soon as new vouchers are added or when there are discounts.