From data to the app people use
Before I wrote applications for a living, I read a lot of spreadsheets. Data analysis at Jukin Media, Power BI dashboards as a DBA, cold surveys and reconciliation work at university. Different jobs, same underlying question: what is this data actually telling us, and who needs to act on it?
When I moved fully into engineering (Node.js, TypeScript, React), that question came along. It's easy to build an app that stores data. It's much harder to build one that surfaces the right data at the right moment, in a form a busy person can trust at a glance.
So the way I think about a feature usually starts at the far end of the pipeline:
- What decision is someone trying to make?
- What's the smallest, clearest view that helps them make it?
- Only then: what data, what API, what schema?
Working backwards from the human keeps the software honest. It's a habit I picked up staring at KPIs, and it hasn't let me down since.