Define what you want the review to produce
A useful weekly review ends with an updated understanding of a company, not a requirement to make a trade. Pick one company and one question you want to revisit. AMD’s free page is a practical place to learn the process; NVIDIA and Palantir are also available as public examples.
Keep a note outside the product containing your previous observation, its source and the condition that would change your interpretation. If this is your first review, create a baseline rather than pretending to have a prior conclusion. You will use it to distinguish genuinely new information from familiar background next time.
Check the assessment and price dates separately
Open Current key factors and read the review date. Then look at the date attached to the price trend. Company factors are scheduled for weekly review, while other sources and observations have their own update cycles. A fresh page visit does not mean every item was refreshed at that moment.
Read any notice about newer filings, insufficient evidence or an incomplete refresh. If an assessment is awaiting review, record that limitation. You can still inspect the company’s disclosures, but you should not describe the retained assessment as a complete account of the newest developments.
Read Since the previous review carefully
Where a verified comparison is available, use Since the previous review to see what changed in the assessment. Open the supporting detail rather than relying on a short change label alone. A factor may have new evidence, while another remains materially unchanged within the collected source set.
An unchanged assessment does not establish that nothing happened anywhere in the company. It means the review did not establish a supported material change in the material it used. Similarly, a baseline or unavailable comparison is not an adverse business development. It tells you what comparison the product can currently support.
If follow-ups on earlier watch points are available, read their status and evidence. Distinguish a checkpoint that has been answered from one with additional information and one still pending. A pending question should remain open in your own notes.
Inspect the evidence behind one important change
Choose the change most relevant to your original question. Open its source and confirm the company, publication date and applicable period. Read What happened separately from Why it matters. The first should describe evidence; the second is an assessment of its possible significance.
Consider an alternative explanation or a missing piece of information. A new announcement can be relevant without proving a revenue effect, and a favorable business observation can coexist with uncertainty elsewhere. The exercise is to identify what the evidence changes, not to force every update into a positive or negative verdict.
Use the timeline to look for later disclosures that may qualify the assessment. When sources disagree, retain the distinction and dates. Do not average incompatible claims into a single confident conclusion.
A sample review note you can reuse
Use a short external note with four fields: baseline, new evidence, interpretation and next checkpoint. The baseline records what you understood at the previous review and which source supported it. New evidence names only material you have actually checked since then. Interpretation states what that material changes, including any limits. The checkpoint specifies what could answer the remaining question.
Consider a hypothetical company that previously described a planned product launch. At the next review, a source confirms that the product is available. Your updated fact is availability; it is not automatically strong customer adoption, recognized revenue or improved profitability. Those are separate questions that may need different evidence. A disciplined note can mark the original launch checkpoint as answered while leaving the commercial questions open.
If nothing relevant has changed, write that no new verified answer was found in the material you reviewed. Avoid the broader claim that nothing happened. This wording preserves the scope of your work and makes the next review easier to compare.
The routine should scale with your available time. Checking one important source carefully is more useful than skimming every company and recording unsupported conclusions. If your watchlist grows, prioritize unanswered questions and new disclosures rather than trying to give every ticker equal attention each week.
Run a focused review when you have limited time
Try an illustrative twenty-minute session rather than an open-ended scan. Spend the first few minutes reading your previous note and choosing the single question that matters most. Next, check dates and visible update notices. Use the largest part of the session to open one relevant source, then reserve a few minutes to record what changed and what remains unresolved. The time allocation is a suggestion, not a product feature or a required schedule.
If the source requires more time, carry it forward explicitly. Write the document title, the section you reached and the particular statement still needing verification. This gives the next session a clear starting point. Do not mark the checkpoint complete merely because you used all the time you had allocated; completion depends on the evidence you actually checked.
At the end, distinguish three possible outcomes: a supported change, a supported confirmation of an earlier observation, or no verified answer in the material reviewed. Each is useful. The routine becomes less reliable when every session is expected to produce a dramatic new thesis, because that expectation encourages small or unrelated developments to be overstated.
Keep an archive of what would change your view
Maintain a dated record outside ThesisMemo with one line for each important checkpoint. Beside the question, state what observation would strengthen the interpretation and what would weaken it. For a hypothetical product launch, availability might answer an operational milestone, while customer adoption would remain a separate commercial question. Avoid combining both into a single pass-or-fail label.
When the next review arrives, read those conditions before looking for new headlines. This order helps you evaluate developments against a question you established earlier instead of inventing a new explanation around whichever story looks most striking. If your original checkpoint was poorly defined, revise it openly and explain why. An improved question is a legitimate result of the research process.
Once a month, inspect the usefulness of the routine itself. Which saved questions led to concrete source checks? Which were too broad to answer? Remove redundant tasks and carry forward the unresolved questions that still matter. You are improving the quality of attention, not measuring success by the number of pages opened. Keep this process separate from any trading decision or expectation of a particular return.
Write a short review note and set the next checkpoint
Finish with three lines in your own document: what changed, what did not change, and what you still need to learn. Attach the strongest source URL to the first line. Keep the final line specific enough that you could recognize an answer in a later disclosure.
A recurring reminder in your own calendar can help you return to the same process. This guide does not assume ThesisMemo creates that reminder or schedules a personal report for you. A saved watchlist simply provides an easy route back to the companies you follow.
On your next review, start with the previous checkpoint. If no supported new evidence is available, say so. Maintaining a clear unanswered question is more useful than inventing an update to justify the time spent reviewing.
Product references
Questions & answers
Does a weekly review require a weekly trade?
No. The objective is to update your understanding and evidence record. A review can reasonably conclude that a question is still unanswered or that the available information does not justify changing your interpretation.
Does a new review date mean the factors changed?
Not necessarily. The product can review the available source set and retain an assessment. Read the comparison and supporting evidence to distinguish a changed factor from a later check of similar material.
What does a baseline comparison mean?
A baseline is the starting point for a later comparison. It does not imply a positive or negative business result. Save your own initial question and return after another successful review if you want to examine what changed.
What if a watch point remains pending?
Keep it open and identify what evidence would resolve it. Pending means a verified answer has not been established in the reviewed material; it is not proof that the underlying event failed to occur.
Do I need membership to learn this routine?
No. You can practice with NVIDIA, Palantir or AMD. Membership is relevant when you need the broader covered-company workspace. Learn the reading process first, then evaluate access based on your actual research needs.