I'm a PM with a few years of experience in the e-commerce and DTC space, and I've recently been inspired to create my own apps given the rise of new AI tools. I've started outlining features, but I soon realized that turning these ideas into a real app is a completely different ballgame. While I'm comfortable writing user stories, I struggle once I'm in the IDE and often find myself second-guessing my decisions, especially when it comes to data modeling.
I've experimented a bit with dbdiagram, and it's been somewhat helpful, but I'm looking to learn how others approach data modeling—particularly those who have non-traditional or PM backgrounds like mine. How much structure should one have before even getting real users? Do you prefer mapping things out visually, or do you work in reverse, starting with user authentication? I really want to understand how other people tackle this process!
1 Answer
When I start building out data models, I focus on the relationships between different data types first. For example, think about a structure like Reddit: you have Users who can create multiple Posts within a subreddit, and each Post can have comments that might be upvoted or downvoted. This helps clear up the necessary data points and how they link together!
Thanks for that insight! I think I'm struggling with the complexity of it all, though. I got stuck diving deep into 'has many' relationships and started to feel a bit overwhelmed. Maybe I'm just overthinking this!