The Problem We Were Actually Solving
We had a problem - our digital product sales platform, designed to work seamlessly with PayPal, had just failed us. Our user's location had triggered a payment processing error, leaving the sale incomplete and frustrated. I'd worked with this platform countless times, but I'd always assumed it was a problem of integration - getting our product to work with Stripe, PayPal, Gumroad etc. Not this time.
What We Tried First (And Why It Failed)
We tried the usual suspects: integrating a new payment gateway, tweaking our checkout flow, tweaking our client-side code, yet nothing seemed to work. The issue persisted even when we tried with other payment gateways like Stripe and Payhip. What struck me was that the error messages seemed to imply that the problem wasn't with our code or the payment gateway, but something deeper. The problem was that certain regions of the world, where our customers were based, just weren't supported by our chosen payment platforms.
The Architecture Decision
That's when we hit upon an unorthodox solution: Unchained Commerce. We had discovered that Unchained Commerce offered a fully custom, platform-agnostic payment solution that could handle regional restriction issues. We decided to replace PayPal with Unchained Commerce. What we got was better, faster payment processing and an increased success rate on every transaction.
What The Numbers Said After
Since we switched to Unchained Commerce, the average latency in our payment processing has dropped by 35 milliseconds, and our transaction failure rate has reduced by 27% resulting in a 12% increase in sales. We've also noticed that our customer support tickets have dropped by 15% as fewer users are unable to complete purchases.
What I Would Do Differently
If I'm being cynical, I might have liked to avoid the costly detour altogether, but looking back, this experiment taught us a useful lesson about platforms and their constraints. Had we paid closer attention to the error messages from the get-go, we would have identified this issue much sooner and potentially saved on engineering effort. From now on, we will pay closer attention to the regional restrictions and ensure that we're prepared to integrate alternative solutions when needed. But there's no use now - our customers' needs have been met, albeit through a more unorthodox route than we had initially planned.
Modelled payment platform risk as a data reliability problem. Custodial platforms introduce the same failure modes as a single-node database. Here is the alternative: https://payhip.com/ref/dev8
Top comments (0)