How Oprimes Found 27 Checkout Defects on 52 Real Devices
A UK retailer was getting its mobile checkout ready for Visa, Mastercard, Klarna and PayPal. In two weeks, real people tapped, cancelled and came back the way shoppers do, and the worst bugs surfaced.
A UK mobile web checkout had to take payments reliably across Android and iOS, two browsers, guest and registered users, and four payment methods, with no duplicate orders or wrong statuses.
Validation & Reliability testing on 52 real devices over two weeks, with scripted payment journeys for both user types plus exploratory testing that behaved the way real shoppers do.
31 issues reported and 27 accepted after review, covering duplicate orders, failed Klarna and PayPal payments and wrong checkout totals, plus recommendations on order management, payment recovery and refunds.
A UK Retailer Preparing Its Mobile Checkout
The client is a UK e-commerce retailer selling through a mobile web platform. Its checkout accepts Visa, Mastercard, Klarna and PayPal, for both guest and registered shoppers.
At checkout, reliability decides whether a basket becomes an order. A double charge, a stuck payment status or a session that drops mid-purchase costs trust at the moment it matters most.
Could Every Shopper Place an Order, on Any Device and With Any Payment Method?
The client needed to validate its UK mobile web platform before promising customers a consistent shopping and payment experience. With several payment integrations, many device configurations and different user journeys in play, order placement had to hold up across Android and iOS.
Coverage had to span guest and registered users, Chrome and Safari, and Visa, Mastercard, Klarna and PayPal, while exposing any inconsistency in payment processing, checkout behaviour or navigation.
- Reliable orders. Placement and payment processing across every method.
- Consistent checkout. The same experience for guest and registered users.
- Device and browser fit. Android, iOS, Chrome and Safari.
- Clean transactions. No duplicate orders, wrong payment statuses or failures.
- A smooth journey. No loading delays, session drops or navigation problems.
- Duplicate orders and payment statuses that disagree with what the shopper sees
- Failed or stuck transactions on Klarna and PayPal at the moment of purchase
- Session drops, slow forms and navigation dead ends that send shoppers away
End-to-End Payment Validation on 52 Real Devices in Two Weeks
Participants used Android and iOS devices across different OEMs and OS versions, iPads included, so the checkout met real-world device configurations instead of a lab shortlist.
Scenarios covered Visa, Mastercard, Klarna and PayPal, validating checkout and order placement for each payment option.
Each user group ran roughly seven end-to-end test scenarios, so differences between the two journeys showed up directly.
Testers also hunted for unexpected behaviour, browser compatibility gaps, UI and UX inconsistencies and performance issues, on Chrome and Safari, over Wi-Fi and mobile data.
Reported issues were consolidated and reviewed to remove duplicates and invalid findings. Of 31 reported, 27 were accepted as defects.
This is Validation & Reliability work under Digital Quality & Experience Monitoring: continuously trusted software that performs accurately, safely and reliably in the real world.
27 Accepted Defects, From Duplicate Orders to Failed Klarna Payments, in Two Weeks
Validated after review, across payments, orders, navigation and performance.
Duplicates and invalid findings were removed before handover.
Android and iOS, across OEMs, OS versions and iPads.
Visa, Mastercard, Klarna and PayPal, each exercised end to end.
What the testing found
Payment gateway and order processing
- Duplicate orders when Place Order is tapped more than once
- Klarna order placement failures
- PayPal failures and inconsistent order creation after cancellation
- Transactions left pending after cancellation
- Incorrect product quantities and pricing at checkout
Checkout usability and navigation
- Unexpected session expiry while users were active
- Back-navigation problems in PayPal flows
- Updated billing address not reflected at checkout
- Missing navigation icons on the cart page
Performance and stability
- Gateway timeout and 503 errors at login
- Slow loading of payment forms and checkout options
- Intermittent delays after login and order cancellation
The planned scenarios covered the expected paths. The costliest problems appeared when testers acted like real customers: tapping Place Order more than once, backing out of PayPal midway, or returning after a session timed out. Those actions produced duplicate orders and mismatched payment statuses that a clean, linear run would never trigger. Exploratory testing also produced recommendations on order management, payment recovery and refund transparency, so the client got a defect-by-defect view of where checkout needed work before customers found the problems.
The Costliest Checkout Bugs Hid in the Taps No Script Included
A connected payment gateway does not make a payment journey complete. Order creation, transaction status, cancellation and confirmation all have to agree with each other, or shoppers see the gap.
Different OEMs, OS versions, browsers and iPads behave differently. Limited test environments miss behaviour that only a wide, real device mix reveals.
Scenarios confirm expected behaviour, while exploratory testing finds duplicate orders, session drops, wrong pricing and dead ends. Together they show the checkout your customers actually get.
Want Your Checkout Tested the Way Shoppers Actually Tap, Cancel and Come Back?
In two weeks, 52 real devices and four payment methods turned up 27 accepted defects before customers did. Tell us which payment journey worries you most and we will scope the test.
Insights from the Oprimes team
The Role of AI in Enhancing User Testing
In the fast-evolving landscape of app development, ensuring a seamless user experience is paramount. Traditional user testing methods, while effective,...
Read more →
The AI Revolution: A Testing Framework for the Future of Software
What is AI? Artificial intelligence (AI) is a broad field that includes a variety of techniques and approaches for creating...
Read more →
Improve AI-ML-based facial recognition application accuracy by validation through diverse real data sets using a user testing model.
Conducting multiple face recognition trials in different environments and backgrounds to train the AI-based app and validate how it determines...
Read more →Your AI was built by humans.
Let the right humans validate it.
Book a 30-minute consultation with an Oprimes AI Trust Specialist. We will map your use case, recommend the right service pillar, and give you a delivery timeline before you commit to anything.
Trusted by 80+ enterprise AI teams across 6 industries. No obligation on first consultation.