Build a repeatable stock research checklist with AI

Good research is repeatable. A checklist ensures that you ask the same core questions about every company, so that you compare them fairly and are less swayed by a persuasive story or a recent price move. AI can help you work through the checklist quickly, but it works best when the structure is yours. This article proposes a six-part checklist, shows how to use an assistant at each step, and explains how to keep a record so that you learn from your own decisions.

Why use a checklist

Investors, like pilots and surgeons, make better decisions when they follow a consistent process. A checklist:

• Reduces the chance of skipping an important question when you are excited about an idea.
• Makes companies comparable, because each is judged on the same dimensions.
• Slows you down at the moment when impulse is strongest.
• Creates a record you can review later.

The six sections

1. The business

• What does the company sell, to whom, and how does it make money?
• What are its main products, segments and geographies?
• Who are its competitors, and what advantage does it have, if any?
• How much of revenue depends on a few customers or products?

With AI: ask for a plain-language description of the business model based on the latest annual report, and for a list of the main segments with their share of revenue.

2. Growth

• How have revenue and earnings changed over the last five to ten years?
• Is growth organic or driven by acquisitions?
• What are the drivers, and are they likely to continue?

With AI: ask for a table of revenue and earnings per share by year, with year-on-year growth, and for management's explanation of the drivers.

3. Profitability and returns

• Gross, operating and net margins, and their trends.
• Return on equity and return on invested capital.
• Free cash flow and how it compares with profit.

With AI: ask it to calculate and tabulate these, then verify them yourself.

4. The balance sheet

• Debt, cash and net debt, and how they compare with earnings.
• Interest coverage and when debts mature.
• The quality of assets, including goodwill.

With AI: ask it to list borrowings with maturities and interest rates, and to highlight any off-balance-sheet obligations.

5. Valuation

• Price to earnings, price to free cash flow and enterprise value to EBITDA.
• How these compare with peers and with the company's own history.
• What the current price implies about future growth.

With AI: ask for peer comparisons, and for the assumptions needed to justify the current price.

6. Risks

• Competition, technology change and regulation.
• Customer or supplier concentration.
• Management and governance, including insider ownership and pay.
• What would make you wrong?

With AI: ask for the main risk factors from the filing, and for the strongest argument against the investment.

A worked sequence

1. Create a template with the six headings and a few questions under each.
2. Collect the latest annual report, the most recent quarterly report and the latest earnings call transcript.
3. Ask the assistant to fill in the first draft of each section, citing where each fact comes from.
4. Check the numbers that matter against the filings, and recalculate the key ratios.
5. Add your own judgement: what you find convincing, what worries you, and what you do not understand yet.
6. Write a short conclusion: the thesis in a few sentences, the key risks, and what would make you change your mind.
7. Decide how much, if anything, to invest, and in the context of your whole portfolio.

Keep notes and review them

Store each checklist in the same place and format, with the date. Include the price at the time and your reasons. Every six or twelve months, revisit past entries and ask what you got right and wrong, and why. This is one of the most effective ways to improve as an investor, because it tests your reasoning against what actually happened.

Adapt the checklist

Different types of company need different questions. Banks need questions about capital and loan quality, property companies about occupancy and debt, early-stage companies about cash runway. Add or change items as you learn, but keep the core consistent.

Common mistakes

• Filling in the template with AI output and not verifying it.
• Skipping the risk section because you like the story.
• Changing the criteria to fit a company you already like.
• Not recording your reasoning, so you cannot learn from it.
• Treating a completed checklist as a prediction.

A short checklist for the checklist

• I use the same structure for every company.
• I verify the numbers that drive my view.
• I always include the case against.
• I record my reasoning and review it later.

The checklist does not make decisions for you. It makes sure that when you decide, you have asked the right questions.