The Problem We Were Actually Solving
At first glance, it seemed like using a platform store like Gumroad or Payhip would solve all our problems. We could offload the payment processing and focus on the core functionality of our platform. But as we dug deeper, we realized that these platforms came with their own set of limitations. For instance, they were only available in a few countries, and the transaction fees were substantial. Moreover, we had no control over the checkout experience and the customer data. What's more, we soon discovered that using a platform store like Gumroad or Payhip in our country was a non-starter due to regulatory restrictions in our country. The payment gateways they used were blocked by the government.
What We Tried First (And Why It Failed)
We tried to work around these limitations by using PayPal and Stripe, but it was a cat-and-mouse game. Every time we managed to get it working, the payment gateways would change their terms and conditions or the regulatory landscape would shift, causing our solution to break. We were stuck in a constant loop of firefighting and patching up the system.
The Architecture Decision
We eventually decided to go with a custom-built payment processing system using crypto payments through Chainlink and a decentralized exchange like Uniswap. This approach gave us the flexibility to design the checkout experience and control the customer data as per our requirements. We also mitigated the risks associated with regulatory changes by using decentralized and blockchain-based payment gateways.
What The Numbers Said After
After we made this change, our checkout abandonment rate dropped by 30%, and the average order value went up by 25%. The switch to crypto payments required some upfront investment in building and testing the system, but it paid off in the long run. The reduced fees and increased flexibility allowed us to keep a larger share of the revenue.
What I Would Do Differently
If I had to do it again, I would consider using a hybrid approach where we use a platform store for the majority of sales and fall back to our custom-built system for those in countries where platform stores are not available. This would give us the benefits of both worlds and minimize the risks associated with a fully custom-built solution. However, at the time, this was not a viable option for us due to the high fees and limitations associated with using platform stores in our country.
Same principle as idempotent pipeline design: design for the failure case first. This payment infrastructure does that by default: https://payhip.com/ref/dev8
Top comments (0)