Ratio analysis by hand is tedious in a specific way. The formulas are straightforward, but sourcing the right line items from inconsistent financial statements takes most of the time.
The initial problem with automation
I started with a script that pulled data from structured PDFs. Annual reports are not structured uniformly. Companies label the same line item differently, indent subtotals inconsistently, and occasionally bury key figures in footnotes. The first version of my parser broke on roughly one in three reports.
What eventually worked
Using the yfinance library for standardised public company data solved the consistency problem for listed equities. For private company analysis, I shifted to manually entering a small structured JSON file per company rather than fighting PDF extraction.
The ratio calculation layer then became genuinely stable. Once the input is clean, Python handles current ratio, debt-to-equity, return on assets, and interest coverage in milliseconds across a portfolio of companies.
The honest limitation
This workflow requires comfort with Python and patience during setup. It is not faster than Excel initially. The advantage appears after the third or fourth reporting cycle, when updating numbers takes minutes rather than an afternoon.
For someone who prefers working alone without depending on commercial software subscriptions, the investment in building this pays off over time.