Automating Donor Philanthropy

These approaches are peculiar to my background and purpose, but the software is customizable through user modification of input parameters.

Phase 1: Form 990 in Excel

I wrote this code in 2025 with data downloaded from government websites and processed with Excel VBA code. That software is accessible on GitHub. Click this link to go to github.com/royrada/Automating-Philanthropy for my GitHub site of Excel VBA code to score IRS Form 990s.

Two local descriptions are

narrative and
README.

Phase 2: Webscraper

A key issue in funding an endowment is the amount of money required. My Python webscraper goes to the website of 501c3 entities that are suggested by the Phase 1 software and looks for quotes that specify the amount of money required to endow a fund. That software is accessible on GitHub. Click this link to go to github.com/royrada/min-endow-scraper and get my Python web scraper to discover endowment limits.

Two local descriptions are

README and
Design.

Phase 3: IRS Master File

A challenge with the Phase 1 system was classifying the entities based on their mission description on their Form 990. However, the IRS Business Master File does the hard work of classification with NTEE codes. This Phase 3 upgrades Phase 1 in several ways, including that the system uses a SQL RDBMS and Python rather than Excel workbooks and VBA. First the BMF is downloaded and used to classify entities, then the Form 990s are downloaded and filtered to pick the best ones according to user definable criteria.

That software is accessible on GitHub. Click this link to go to github.com/royrada/filter-IRS-990-BMF and access the Python and RDBMs system that in a user-customizable pipeline downloads IRS data, imports the data into the RDBM system, classifies the 501c3 entities, and rank orders them.

Three local descriptions are