I ran a non-profit website in Brazil where schools could look up a question-by-question breakdown of how their students did in the high-school exit exam. I was

I ran a non-profit website in Brazil where schools could look up a question-by-question breakdown of how their students did in the high-school exit exam. I was a bit flummoxed by the process of maintaining a decently sized database with that data, and realized that there were only so many visualizations I'd need for it. So, over a week, I just had an algorithm compute everything I would need for each school for the several million students in the db, and then handily save a file for each institution with that data in an S3 bucket. No databases, no complex logic, just a few pennies of cost for S3... and it made building out the visualization such a breeze!