Ask why revenue, margin,
labor cost changed.
Get grounded answers with page-level citations from SEC 10-K and 10-Q filings in seconds. Stop searching PDFs. Start explaining variance.
Why Financial Metrics Changed
Financial variance analysis often requires analysts to manually search SEC 10-K and 10-Q filings, read Management Discussion & Analysis (MD&A), and connect changes in financial metrics to the underlying business drivers. Slow, error-prone - and it doesn't scale across 7 companies and 4 industries.
- The manual crawlOpen the filing, search the filer, find the number, read the MD&A context.
- The trust gapNo guarantee the paragraph you landed on is the one that moved the metric.
- The scale wallDo this across a basket of companies and the process breaks down completely.
█
How AI Analyzes SEC Filings
Wage inflation and CA minimum wage increases, offset by sales leverage. Cited: CMG 10-K p.34, 10-Q p.12.
Accuracy on Real SEC Filings
Tested against 40 real questions with known correct answers, before and after adding a relevance-ranking step. The hardest cases went from being missed entirely to found first.
Real questions, real answers.
- 01
Explains financial variance
Digs past the headline number to the drivers - price, volume, mix, wage inflation, FX - so you can explain why any metric moved.
- 02
Searches SEC filings
Reads straight from 10-K and 10-Q filings across 7 companies - no more manual searching through PDFs.
- 03
Finds supporting evidence
Retrieves the exact MD&A passage that backs each claim, ranked for relevance before anything is written.
- 04
Cites every answer
Each claim points to the filing and page, so you can verify it before it goes in a report or a deck.
- 05
Works across industries
Indexed on restaurant, retail, healthcare, and energy filings with no drop in retrieval accuracy between them.
Frequently Asked Questions
What is financial variance analysis?+
How do analysts use SEC filings?+
What is MD&A?+
Can AI analyze 10-K filings?+
Does this work on 10-Q reports?+
Which companies are supported?+
Does it run locally?+
Running in two minutes.
Want to run it yourself or see exactly how it works under the hood? Here's the full setup.
# Clone & setup git clone https://github.com/redsandr/rag-variance-explainer cd rag-variance-explainer python -m venv venv && source venv/bin/activate pip install -r requirements.txt cp .env.example .env # Build index & run python src/build_index.py streamlit run app.py
What's next.
- 01International filings, not just US-based companies
- 02Follow-up questions in the same conversation, not just one-off queries
- 03One-click setup, no command line required
- 04A version tuned specifically on financial language for even higher accuracy
- 05A public demo anyone can try without installing anything
Like what you see?
If this project helped you, consider starring it on GitHub. It helps others find it too.
Star on GitHub