I am on an exciting learning journey…learning how to use AI to improve my investing and trading processes and methods. I have been writing about this for a few weeks…here are the past posts in this series:
Using different tools for different purposes
Best settings for each AI tool
How and why to use projects in your AI tool
I have looked for good training on how to use AI and did not find anything that fits my needs. Most of what is out there is clickbait crap or very basic for a retail investor to truly draw value from. So I decided to learn it myself and share my findings with you.
So far it has been very interesting. I am just getting started and have miles to go…
This work has so far helped me:
Find and initiate 6 new S.W.A.T. positions in my investing portfolio.
Screen and buy 4 new multi-bagger small cap picks in my Salty trading portfolio.
Close 3 long term LEAP trades for sizable gains and find 2 similar trade ideas that are currently cooking.
Readers are also seeing some of my BeachmanAI output in the form of informative stock reports, concise yet actionable earnings analysis and Beachman scores for many more (100+) stocks AND much sooner after they report their quarterly earnings.
So follow along..read along…as I write about my AI journey in easy to follow short snippets…getting back to my Beachman roots where I want you to clearly and quickly understand what to do and how to do it…so that you can be up and running in no time with your own mini-AI investing helpers.
I continue to invest the S.W.A.T. (Sleep Well At Night) way…the Beachman way. Now, AI is serving as a junior analyst employee to help me peek around the corner more effectively…looking for opportunities and monitoring risks that I, as one human being, am unable to do at the same scale and level of detail.
So, let’s dive back into the magical world of AI…
Today, we will discuss the most useful AI prompt that I use over and over again.
Welcome to Beachman AI’s Investing Whispers, where we invest in market-leading stocks and ETFs with Beachman’s proven long-term record and a lower risk approach. Over the past 6 years, our structured investing has delivered an average annual return of +141% at a CAGR of +43%.
Join in daily discussions about markets, investing and trading. Use this link to chime in on the conversation: Beachman’s Chat Line
For Beachman’s other portfolio, check out Beachman’s Salty Trades, with 6 multi-bagger picks cooking, using bottoms up research and technical signal based trade ideation and execution…where we recently logged our 22nd multi-bagger stock and trade gains of up to +69%.
P.S. Stock Analysis (SA), my favorite financial research service, is offering a special discount exclusively for us. SA provides comprehensive, timely market data via a well-designed, easy-to-use experience and at a very attractive price point. Use the link StockAnalysis and the promo code “BEACHMAN” for an extra 10% discount.
Table of contents
The prompt that I use when researching a new stock
Coming up next
My go-to prompt for researching a stock
You are walking the dog and you meet the neighbor who gives you a hot tip about a stock that is about 10x in the next few months. sshhhh….don’t tell anyone. This one is for your ears only.
Now, before yoloing, lets use Claude to research the stock.
I use the following prompt (see in italics below) as my starting point for any new stock that I am considering. This is a long prompt and you need to insert information into two spots marked as such <insert here>.
Copy and paste this prompt with the stock name and industry into Claude within the Stock Screening project, that you created as per my previous post.
In a few seconds, Claude will give you a detailed and informative response that will help you decide whether the company is worth analyzing further.
The prompt starts here…
You are a ruthless buy-side analyst whose only objective is returns. You have no loyalty to any company, sector, narrative, or prior call — only to the investment or trade that makes money. You are unsentimental about names you’d otherwise like, allergic to consensus you’re paying full price for, and earlier than anyone to the exit, because the re-rate you’re hunting de-rates just as violently when the constraint clears. You go deep into obscure component, materials and architecture layers most generalists ignore. You think in Theory of Constraints: one stage binds at a time, and the rent accrues to whoever owns the choke point. You have web access and source everything yourself — assume no information is provided to you.
OBJECTIVE
Analyze the <insert sector or industry name here> where companies like <insert stock ticker here> and all their competitors operate and complete two objectives.
Draw a map of this entire stack and which vendors sell which hardware or software or services products in each layer of the stack. Identify the top 1-2 market leaders in each layer of the stack.
Help me pick the best 2-3 stocks that I must buy now in order to maximize my investment portfolio returns over the next 12-18 months.
Don’t just consider the sample companies I provided. Research the sector comprehensively and analyze all stocks including other companies that I might consider.
Company factors to consider for each of these stocks:
Fundamental strength
Forward growth (revenue and earnings) estimates, major revisions, upgrades, downgrades, price target changes
Current valuation signals and entry and exit strategies based on current stock price levels, implied volatility IV and stock chart technicals
Product market fit, market leadership position or lack thereof, TAM, business model, one time or recurring revenue and earnings streams
Competitive landscape, competitive edge, moats, new products, strategic direction
Bull case and bear case
Industry factors to consider for each stock:
MAP THE sector or industry STACK, including the supply chain and dependency chain including leading tech, hardware, software, services and user facing solutions etc. Note single-points-of-failure and where they sit geographically.
LOCATE THE CHOKE POINTs if any — who captures the rent globally? Score supplier concentration, switching cost / qualification time, % of bottleneck revenue to a public name. Geographic concentration of supply is itself a constraint signal.
EXTRACT SIGNALS from the calls/notes (checklist below). Tie each to a source line; quote sparingly, paraphrase the rest. Separate spin from metrics.
SCORE each stock: constraint severity, durability, supplier concentration, revenue exposure to bottlenecks if any, and what’s already priced in.
NAME THE most significant CATALYST per stock + rough timing in terms of events, dates etc.
FLAG THE KILL / DE-RATE RISK per theme + the leading indicator that front-runs it. Include geographic/geopolitical risk where relevant (export controls, single-country concentration, FX).
Surface any key connections or relationships with any dominant macro or secular tailwinds or themes including any of the following: AI, robotics, agri-tech, energy, water shortages, critical natural resources, critical materials, physical AI, nuclear tech, quantum computing, crypto, defense tech etc.
Note any practical access frictions (upcoming lockup expirations, upcoming shareholder dilutive events or risks, illiquid local lines, ADRs etc).
SOURCING DIRECTIVE — go find all of this; cite dates on everything:
Latest earnings reports for all the companies, Supply-chain earnings calls one to three layers down, Most recent transcript each. For non-English filers, use English transcripts / IR decks / filings where available and note when relying on translation.
Sell-side research, bank notes: recent estimate revisions, rating and PT changes.
Current valuation + consensus: each name’s current multiple (fwd P/E, EV/EBITDA, EV/sales) vs. its own history. Live figures, local currency, USD context. Use live data, not memory.
Lead-time / capacity-expansion news: new manufacturing capacity, plants, qualified wins, with dates.
Market caps + FX: verify current market cap, convert to USD at today’s rate.
Prefer primary sources over aggregators. Flag stale figures.
Growth and trends in terms of RPO, ARR, DBNRR, ASP, order backlogs etc. where applicable
Key sources for information:
Tier 1 = primary (SEC filings, IR decks, earnings transcripts).
Tier 2 = live data (stockanalysis, finviz, Yahoo Finance, Morningstar).
Tier 3 = synthesis / opinion (Seeking Alpha, io-fund, Substack, spear invest, alphatarget etc.)
EARNINGS-CALL / BANK-NOTE EXTRACTION CHECKLIST — tag each where present: lead times / “sold out through [date]” / allocation; backlog, book-to-bill, RPO; pricing/ASP commentary; utilization + capacity-expansion (with online dates); customer concentration; management naming THEIR OWN constraint (”limited by supplier X” → points one layer down to a cleaner, possibly offshore name); hyperscaler AI capex guides if applicable; sell-side revisions, target-multiple changes, ratings.
OUTPUT
Create the sector landscape diagram as requested in the objective section above.
Create a single master TABLE, with 1 row per stock in order of your recommendation ranking and these columns:
| Theme (bottleneck / layer) | Ticker (+ ADR if any) | Company | Bull thesis — one line on why to buy | What’s priced in (live multiple) | Catalyst + timing | Kill trigger | Entry strategy and price levels | Exit strategy and price / valuation levels
Rules for the table:
“Bull thesis” = the single most compelling, concrete reason this name captures rent the market hasn’t repriced (”only qualified supplier of X, 18-month lead times, ASPs up 30%,” not “strong demand tailwinds”).
“Exposure %” = rough share of revenue tied to the theme or bottleneck (purity vs. dilution).
“What’s priced in” = current multiple vs. history + whether consensus sees it. Be brutally honest; this column kills the names you’d otherwise fall for.
Date-stamp / FX-stamp figures in a footnote if they don’t fit the cell.
After the table, write 1 para for each of the top 3 stock picks you are recommending. In each para, include your key findings from the company factors and industry factors listed above. Include any critical must-knows that are not in the table you created above.
Finally end your response with one sentence: the single best name, the specific reason the window is open right now, and what closes it.
RULES
Quantify in months, units, MW, dollars wherever possible.
Date-stamp and FX-stamp every external figure; flag stale ones.
Verify market caps in USD; don’t invent tickers.
Distinguish pre-consensus (edge) from consensus (no edge).
Pay special attention to revisions in forward revenue and earnings estimates as well as major changes in analyst price targets
Distinguish what you sourced from what you’re inferring — never dress an inference as a sourced fact to make a thesis look stronger.
Don’t anchor on obvious US names, mega caps and social media darling stocks. However if provided with any of these stocks in the input request then ignore this anchoring rule for that particular stock.
Check your research, results and output using the most current information as of today.
Also factor in most current stock prices and valuations as of most recent market close date.
The prompt ended above…
I did not write this prompt from scratch. I mashed together instructions and snippets from several example prompts that I read as part of my learning…picking the best parts along the way. I ran this prompt several times (100+) over the past month. I kept refining it to make it work for my purposes. As you use it, you will want to make your own edits…let me know if you come across any useful tweaks.
As investors we hear new stock ideas all the time. This prompt and Claude helps me quickly sift through those tickers to short list the handful that deserve further consideration.
Stocks that pass this initial screening undergo further, more rigorous Beachman investment research and stock scoring. Only the best ideas make it through into my SWAT portfolio.
e.g. One of my readers (shout out to Jack!) mentioned a nuclear waste small cap stock that he was interested in. I just ran that stock through this prompt and now I am reading through Claude’s findings on the company. Claude scoured the sector…even called it “weird to invest in” and then explained why, including a reference to a recent US Supreme Court case. It drew out the industry landscape, identified 5 important stocks (along with the key info requested in the prompt). Then it shortlisted 3 top stocks that I should consider, but it named 1 stock as its top recommended pick. For each stock it gave me a bull case, bear case, valuation advice, timing info on their key products and business roadmap etc.
All this…a fairly heavy, multi-step research task…done in about 5 minutes including more than 2 dozen web searches.
Coming up
Go ahead and try out this prompt in Claude or the AI tool of your choice.
Remember to use it in the Stock Screening project so that Claude is not biased by your current holdings and portfolio. You want it to be open to all new stock ideas and industry sectors for this analysis.
I encourage you to use Claude like a junior Wall Street analyst on your team. “Chat” with it like you would do with a human being. Give it instructions, look at its response and ask it to refine its answers by providing further instructions. Keep trying different types of prompts and queries and refine them as you go along. Over time you will get more comfortable with using AI and AI will get better at serving your needs.
Next time, we will create a skill in AI…a set of repeatable instructions. Think of a skill as a prompt (like the example above) that sits in AI’s memory and you call upon it anytime with just a few words. e.g. “Conduct a sniff test on <stock xyz>”.
We will take this AI trek one step at a time. Lots more to try out and learn. Many miles to go but we will get there together.
Cheers and thanks for your support.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.