I’ve been experimenting with a simple but surprisingly powerful method to analyze small-cap stocks using a combination of Seeking Alpha data and ChatGPT. Now I didn’t go into this with any grand expectations. I was just curious: what happens when you combine live fundamental metrics with a LLM?
Turns out, the results are better than I anticipated.
Seeking Alpha Data Source
Seeking Alpha Premium is my favorite platform for updated stock data. The issue with a lot of the tools on the market is they either don’t offer data exports, or you have to spend considerably more than the cost of an SA subscription. Seeking Alpha’s stock screener for Premium members is the best and most economical solution here. You can export valuation metrics, growth figures, profitability, and even SA Quant Scores to feed directly into ChatGPT.
Even though ChatGPT can search the internet for free data, it obviously can't access paywalled content, which is where most of the updated fundamentals on stocks are found. A lot of sites show simple metrics like the PE ratio, but this information is posted outside of any earnings revisions. This why the combo of Seeking Alpha and ChatGPT works so well. I download the most recent metrics, and ChatGPT provides the analysis.
How to Run Stock Analysis in ChatGPT
For this example, I am focusing on USA small-cap stocks. It helps to segment your data in advance since there are so many stocks out there. Here's how I did it:
- Create CSV Exports: After setting up the stock screener for USA small caps, I exported three data files from Seeking Alpha. These are based on the tabs within the screener results. I exported one each for valuation, growth, and profitability metrics. Even though the screener has multiple tabs, you can only export one tab of data at a time.
- Convert to CSV: This might be optional, but I converted each Excel file to CSV format to make it easier for ChatGPT to parse. Not sure if it helped, but I do know the syling of Excel and Word files can affect ChatGPT's ability to read documents in order.
- Begin Chat Session: Before uploading the files, I started the session with a general prompt: “What are some fundamental investing ideas for choosing small cap stocks?” I do this to set the context of the chat so I can see how it plans to analyze the files.
- Upload & Analyze: I uploaded the three CSVs and asked: “Given the small-cap investing strategies you shared above, which stocks would you pick from this dataset?”
For my small cap experiment, ChatGPT suggested a shortlist of promising small caps. From this list I selected Vital Farms (VITL) to dig deeper on. I wanted to see if Seeking Alpha would provide more confirmation that this was indeed an attractive stock from a fundamentals perspective.
Vital Farms
The AI selected Vital Farms based on a combination of solid revenue growth, cash flow, profit margins, and return on equity. I dug a bit deeper by prompting ChatGPT to pull recent analyst ratings and sentiment data across various sites. The output showed that sentiment was overwhelmingly positive, with a 12-month price target in the $45–46 range.
From there, I pivoted back to Seeking Alpha to get even more detailed data. Vital Farms, for those unfamiliar, is in the food production space. They package and sell eggs and butter from pasture-raised animals.
Seeking Alpha's quant ratings graded it a Buy nearly across the board, but I wanted to challenge that a bit. No stock is perfect, so let's look at some of the potential negatives.
Seeking Alpha recently launched their own AI stock summary reports. HEre are the potential risks outlined for Vital Farms:
- Egg prices are elevated due to avian flu and may not be sustainable.
- The high P/E and P/B ratios could indicate overvaluation.
- The absence of a dividend might turn off some investors.
The report also suggested that future gains might simply track the broader market rather than outperform it. The forward P/E is 26, but the sector median is closer to 15.7. Similarly, forward Price-to-Sales was nearly 2.0 vs. a sector median of 1.13.
So even though the fundamentals looked solid, it’s clear there are valuation concerns.
The Real Value of This Approach
What stood out to me was the speed with which ChatGPT highlighted a few promising stocks. That's not to say I would blindly buy into their suggestions, but it was nice to get some fresh ideas from a different perspective.
The key is this process only works if your data source is reliable and updated. If you feed ChatGPT bad data, then it will base decisions off of that incorrect information. Having a data source like Seeking Alpha is important because they update their data daily.
Final Thoughts
The combination of reliable data and AI-driven analysis is definitely worth playing with, even if just for generating stock ideas to research further.
If you want to try this yourself, check out the Seeking Alpha special offers page. Filter through a stock screener and export your own CSVs. From there it's as simple as tweaking your prompts and seeing what stocks they highlight. Perhaps the best way to find hidden stocks is to let the data and robots work together?