I’m pretty particular about my spreadsheets, lol, so I made my own. It had two kinds of columns — raw data (pulled from the CDS, logging the various data that I cared about), and “derived” columns (used the raw data to create calculations I could use for filtering and sorting). For a pedestrian example of a “derived” value, “Admit %” (using the raw applied count and the raw admitted count). For a more involved derived value, “Schools with a higher women’s draw rate than men’s draw rate” (which used several different columns and, in fact, other derived columns as well).
It was a very useful spreadsheet, but every time I used it I was kind of mad that there wasn’t a more universally accessible dataset that others could benefit from. And, also, the knowledge that my data was incomplete, reflecting just the schools I had entered. And, also, the knowledge that in a year there’d be a whole host of new CDS files, and my dataset would be out of date.