PricePoint

First Entry: PROBLEM IDENTIFICATION & PROBLEM VALIDATION. One of the assumptions I carried into building @tryPricePoint was that the biggest problem in social commerce was fraud. It wasn’t an unreasonable assumption. I mean, every other day, someone is posting screenshots of a failed transaction, a fake vendor, or another warning for people to “be careful.” I thought if we could somehow stop fraud, we’d have solved the problem. But somewhere in the middle of finding what was unworkable, urgent, unavoidable about the problem, I was also paying close attention to my own behaviour, I realized something that made me pause. Before buying from an online vendor, I don’t just buy. I instinctively just know to check followers, read comments, search for previous customers, ask friends, and convince myself that this person is probably legitimate. The funny thing is, almost everyone I know does the same. We’ve become so used to it. That was the moment everything changed for me. Fraud wasn’t the problem we were trying to solve at the end of the day. It was merely one consequence of a much bigger one. The real problem is that trust has become a responsibility buyers carry on their own. We have quietly accepted that making your own due diligence is just part of shopping online. That shift in perspective completely changed how we thought about PricePoint. We no longer asked, “How do we stop fraud?” During Problem Validation, we asked a better question: “What would social commerce look like if buyers didn’t have to become investigators before every purchase?” The result led up to one irrefutable conclusion: PricePoint.

When deploying the py architecture for one previous project, meeting security and data guidelines for mobile app audits on Day 1 taught me the value of data design. As I map out the pre-build phase for @tryPricePoint, I am applying some of those infrastructure principles to solve a common BE problem like normalising unstructured data from external platforms. To ensure this app scales smoothly from hundreds to hundreds of thousands of records, the technical methodology will prioritise a decoupled ingestion pipeline