The Problem We Were Actually Solving
Our creators were using platforms like WhatsApp and Facebook to sell their digital products, but those platforms were taking a huge cut of the revenue. On top of that, the platforms were also taking charge of payment processing, which meant our creators had to deal with all the complexity and fees associated with that. We knew we had to create a platform that would allow our creators to sell directly to customers, keeping all the revenue and control.
What We Tried First (And Why It Failed)
We thought using an API-based integration with an existing platform store would be the way to go. We tried integrating with Gumroad and Stripe, but it was a nightmare. The fees were too high, the control was too limited, and the support was non-existent for our specific use case. We even tried using a headless e-commerce solution, but it was too complex and required infrastructure we just didn't have. In the end, we had to admit that the platforms we were trying to use just weren't designed for our needs.
The Architecture Decision
We decided to build our own storefront from scratch, using a combination of Node.js and PostgreSQL on the backend, and a React frontend. We knew it would be a lot of work, but we also knew it would give us the flexibility and control we needed to succeed. We also opted for a custom payment processing solution using Square, which allowed us to keep the revenue and control we needed.
What The Numbers Said After
After launching our storefront, we saw a massive increase in revenue for our creators. We reduced the fees from 20% to less than 5%, and we were able to support over 100 payment methods, including mobile money. Our creators were able to set their own prices and control their own sales, which gave them the flexibility they needed to succeed. And with our custom payment processing solution, we were able to support transactions in real-time, reducing latency and increasing overall performance.
What I Would Do Differently
If I had to do it again, I would invest more time in testing and validating our architecture before launch. We underestimated the complexity of payment processing and had to add a lot of extra work to get it right. I would also invest more time in API design and documentation, as this was one area where the platform stores really fell short. And finally, I would consider using a serverless architecture for the frontend, as this would have given us more flexibility and scalability in the long run.
In the end, building a custom storefront for our creators in Ghana was a huge success. But it wasn't easy, and it wasn't cheap. It required us to take a step back and re-evaluate our architecture, our technology stack, and our partnership strategy. And it required us to be willing to take some risks and invest in the right solutions to get the job done.
Top comments (0)