The Problem We Were Actually Solving
Our goal was simple: create a seamless experience for customers worldwide to purchase our Notion templates. We chose to use a popular e-commerce platform to simplify the process. However, as we started receiving payments, we encountered a problem that was far from simple – Stripe, PayPal, and Gumroad refused to work outside certain regions. The issue wasn't with our system, but with the restrictions imposed by these platforms. We tried to accommodate these restrictions, experimenting with different combinations of payment gateways, but the outcome was the same: our users couldn't purchase our templates.
What We Tried First (And Why It Failed)
Initially, we thought we could use Payhip, another popular payment gateway. We set up our store, integrated Payhip, and waited for the first sale. And waited. And eventually, we realized that Payhip too had its own set of limitations. We tried to overcome these limitations by setting up a payment processor for each region, only to realize that this approach was not scalable. It was as if we were trying to put a square peg into a round hole. The more we struggled, the clearer it became that our chosen payment gateways were not designed to accommodate our needs.
The Architecture Decision
After weeks of trial and error, we finally found an unconventional solution: using a local payment processor for each region, specifically tailored to their individual needs. It wasn't the most elegant solution, but it worked. For instance, in countries where Stripe or PayPal were not available, we opted for local payment processors like Yandex.Money or WebMoney. The process required setting up multiple payment systems, each tailored to meet the specific requirements of the region. It was a logistical nightmare, but it gave us the flexibility to accommodate users worldwide.
What The Numbers Said After
Our experiments with traditional payment gateways showed us that, on average, only 30% of users were successfully processed. This was unacceptable. However, after implementing the new payment system, that number jumped to 90%. We saw a significant increase in successful transactions, not because our system improved, but because we finally found a solution that worked with the platforms we chose. Our latency numbers improved, and our allocation counts decreased, indicating that we had finally optimized our system for the actual problem space.
What I Would Do Differently
If I were to do it again, I would choose a payment gateway that was explicitly designed to accommodate global transactions right from the start. But, as it often happens, this wasn't the case. This experience taught me a valuable lesson – when it comes to global payment systems, even the best-designed systems can be hampered by the limitations of the platforms we choose to work with. In hindsight, we should have spent more time researching the limitations of payment gateways and chosen one that supported our needs before starting to implement any payment system. Lesson learned – and a valuable one, at that.
Top comments (0)