Research practice
Investment Thesis Tracking Template: What to Check Next
An investment thesis tracking template helps you remember what you believed, why you believed it and what would make you reconsider. Start with one company, one claim and one next check. A useful record should be easy to update without rewriting the original argument after the outcome is known.
Copy this template into your research notes
| Field | Your entry |
|---|---|
| Company and research date | Name, ticker if applicable, date |
| Claim | What do I think will happen, over what period, and why? |
| Business driver | What would cause that outcome? |
| Observable evidence | Which disclosed facts would help test the driver? |
| Current evidence | Latest observation, reporting period and original source |
| Counterargument | What is the strongest competing explanation? |
| Review trigger | Which event, release or date will prompt another check? |
| Evidence that would weaken the claim | What would make the original explanation less convincing? |
| Unknowns | What have I not established? |
| Update record | What changed, when, and how did it affect my view? |
You can use a document, notebook or spreadsheet. The format matters less than retaining the original claim and linking each update to evidence.
Start with a company you already follow, or use OpenBit's stock research page as your starting point. Keep your completed template in your own notes as you work through the questions below.
Write a claim you can investigate
“This is a great company” gives you little to check. A more useful claim identifies a business mechanism and a timeframe:
Over the next two reporting periods, existing customers will expand their use of the service, supporting subscription revenue growth.
Now ask what you would need to observe. Does the company disclose a suitable customer metric? Can you separate existing-customer expansion from new customers or acquisitions? If the answer is no, mark the limitation. Do not quietly replace an unavailable measure with a convenient but different one.
A filled example
The following company and figures are fictional, included only to demonstrate the method. They are not benchmarks or a trading rule.
| Field | Example research note |
|---|---|
| Company | ExampleCloud, a fictional subscription business |
| Claim | Existing customers will expand spending over the next two reporting periods |
| Proposed evidence | A consistently defined measure of spending by the existing customer group |
| Starting observation | Revenue from the starting customer group is 104% of that group's revenue in the prior comparison period; this fictional measure excludes new customers |
| Counterargument | Total growth could come from new customers while existing customers reduce spending |
| Next observation | The same measure is reported at 97%; verify that its definition and measurement window are comparable |
| Interpretation | If comparable, this weakens the customer-expansion claim; total revenue growth alone does not resolve the concern |
| Next check | Read the explanation, identify any definition changes, and record what remains unknown |
The purpose is to show your reasoning changing in response to evidence. It is not to turn one metric into an automatic buy or sell instruction.
Find evidence that matches the claim
For companies that file them, the annual 10-K and quarterly 10-Q are useful starting points. The business description, risk factors and management discussion help locate the company's explanation and relevant uncertainties; financial statements provide a separate set of figures to examine. Investor.gov: How to Read a 10-K/10-Q
A Form 8-K can report significant events between periodic reports. Check whether an event is relevant to your specific claim before adding it to your research log. Investor.gov: How to Read an 8-K
Keep management's explanation separate from your interpretation. For every important observation, save a source link, the period it describes and the date you reviewed it. If the data is not available, “unknown” is a valid entry.
Review without moving the goalposts
At the next review, compare the new evidence with the original claim. Use four plain labels: supported, weakened, unresolved, or needs a revised question. Explain the label in one or two sentences.
Then add a dated update instead of overwriting your starting view. If your question changes, preserve the old version and state why the new one is more useful. A changing share price can prompt a review, but it does not by itself explain whether the business mechanism occurred.
Your next research question should come from the largest unresolved assumption. For example: “What evidence would show that existing-customer growth is slowing?” That gives your next reading session a purpose—and makes it possible to check later whether you actually answered the question.
Take your next question to OpenBit
Open OpenBit's research question page with one company and one unresolved claim in mind. Use the question you wrote in the template to guide the session. Check the sources and coverage of any answer you receive, then record what you learned in your own notes.
If a manager's reported holding is part of your evidence, read our 13F filing guide before adding it to the template. Keep the reporting period and the limits of that evidence beside your observation.