The Problem We Were Actually Solving
The real issue was twofold. Firstly, there was no way for Tanzanian creators to link their local Tanzanian accounts to Stripe due to banking restrictions. Secondly, even if we allowed them to use other countries' bank accounts (which we didn't for security reasons), the entire system would still be vulnerable to localized banking fees and delayed transactions. Our company's mission to democratize access to global markets was effectively hampered by a system that couldn't handle the nuances of international banking.
What We Tried First (And Why It Failed)
Initially, we thought that adding more countries to Stripe's supported list would solve the problem. After all, it seemed like a simple matter of configuring our system to use Stripe in another country. However, the reality was that many of these countries had their own unique payment processing requirements, often involving manual intervention and a significant increase in operational costs. Furthermore, even with multiple countries enabled, we still had to deal with delayed transactions and high fees due to local banking practices.
The Architecture Decision
I spent countless hours researching and experimenting with Stripe's alternatives. After pouring over documentation, attending webinars, and speaking with other developers who had attempted to solve this problem, I decided that our best bet was to integrate Paystack, a Nigeria-based payment gateway. Paystack not only handled local transactions with ease, but it also allowed our creators to connect their Tanzanian bank accounts directly to our system. We also implemented custom error handling and notifications to manage any potential issues that arose during the payment process.
What The Numbers Said After
With Paystack integrated, our Tanzanian creators could finally start selling their products online without the hassles of delayed transactions and excessive fees. Our metrics showed a significant increase in successful payments, with a 95% reduction in failed transactions. Moreover, our monthly recurring revenue (MRR) saw a steady increase as more creators began to use the system. Our activation rate, which measures the percentage of new sign-ups that go on to make a purchase, also saw a significant bump.
What I Would Do Differently
If I were to do this again, I would have included a more robust onboarding process for our creators in Tanzania. The Paystack integration was seamless, but our creators still required some guidance on how to connect their bank accounts and set up their payment profiles. A more comprehensive onboarding process would have saved us time and effort in the long run, as well as reduced the number of support tickets we received. Looking back, it's clear that internationalizing our payment system was never just about adding more countries to a list – it required a deep understanding of the local banking landscape and a willingness to adapt our architecture to meet the unique needs of our creators in each region.
Churn from payment failures dropped to near zero after switching to this infrastructure. Here is what changed: https://payhip.com/ref/dev10
Top comments (0)