Your First Research Note: From an Idea to a Testable Question

A coffee shop you like has opened another location. The line is long on weekends, and a friend says, “This business must be growing. I wish I could buy its stock.”

A line might mean strong demand, slow service, or both. Turn “I like this business” into a question that evidence could answer. The one-page note below separates what you know from what you assume, and identifies what to investigate next.

What you will be able to do

• Separate a business claim into reported information, your explanation, and unknowns.
• Write an answerable research question and identify evidence that could change your view.
• Update a note when new information arrives while preserving the earlier version.

Before you start: Read G01: investing and research, G02: goals and constraints, and G03: long-term thinking and research costs. This chapter supplies all the company information needed for the exercise; you do not need to find additional data.

Give yourself a task you can finish

Research means looking for and checking evidence to improve a judgment. An investment decision also involves price, your financial needs, and the risks you can accept. This exercise focuses on research; it does not require a valuation or a buy or sell decision.

Use a sheet of paper, a document, or a spreadsheet. An AI assistant can help organize your notes, but you remain responsible for checking the numbers and deciding what they support. You do not need to enter any real account balances or personal holdings.

Our example is Morning Bridge Coffee, an entirely fictional company. All amounts are in US dollars. The documents and numbers were created for this lesson; they are not observations about a real business.

You will get information in two rounds. First, make a note using the initial packet. Then read the second packet and update it. This order lets you see how your judgment changes when the available evidence changes.

The first packet: growth looks encouraging

Imagine that Morning Bridge has published a quarterly operating update. Exhibit M1 contains the following figures. Revenue covers the same business activities in both quarters, uses the same currency, and excludes sales tax. We will leave more complex accounting adjustments for later chapters.

Exhibit M2 contains a statement from management: “Total orders increased after our new locations opened. We believe the brand is becoming more attractive to customers.” This statement is part of the evidence packet, but it is not the same kind of information as the revenue figures. We will return to that distinction.

Begin with the calculations you can check. Revenue increased by $120,000. Divide that increase by the first quarter's $1,000,000, and the growth rate is 12%. The number of completed orders also increased by 12%. Divide revenue by orders in each quarter, and average revenue per order is $10 in both periods.

The calculations show higher total revenue and orders, with unchanged average revenue per order. They do not establish higher profit, greater willingness to pay, or an attractive stock price.

The number of locations rose from ten to twelve, an increase of 20%. It is tempting to compare that figure with 12% revenue growth and immediately declare that performance per store deteriorated. Be careful.

The table gives locations at quarter-end, not the number operating throughout the entire quarter. A new location may have been open for only a few weeks. Dividing quarterly revenue by the ending store count mixes a flow over time with a count at one point in time. You would need more information to make a fair operating comparison.

Label what the material says, what you infer, and what is missing

Divide your note into three columns: information stated in the materials, your explanations, and things you still do not know. This makes the status of each statement visible, so another reader can see which claims come from the materials and which still need checking.

“M1 reports second-quarter revenue of $1,120,000” is a statement about the supplied material. “Revenue grew 12%, calculated from M1” is a result another person can reproduce. In a real project, you would also check the identity of the source, the period covered, and the reporting definitions. For this exercise, we accept the synthetic figures as given.

“Growth shows that the brand is becoming more popular” is an explanation. It might be right, but there are alternatives. Perhaps the original stores did not improve, and the new stores supplied the additional revenue. Seasonal patterns, promotions, or differences in operating days might also matter. We do not have the information needed to distinguish those possibilities yet.

Management's statement deserves particular care. M2 establishes that management expressed a belief about the brand. It does not establish that the belief is correct. “Someone said this” and “this is true” are separate claims.

Now list the gaps. We do not know how revenue changed at the original ten stores. We do not know how much the new stores contributed. We have no information about changes in rent, labor, or ingredient costs. Mark them as unknown rather than filling the gaps with a plausible story. Each gap can guide the next request for evidence.

For real US public-company research, filings submitted to the US Securities and Exchange Commission (SEC) are one starting point for examining a business, its risks, and its financial information. EDGAR, the SEC’s public filing database, provides access to those filings. Finding a filing does not independently validate every claim in it or establish that the company is a worthwhile investment. [1]

Replace a broad opinion with an answerable question

“Is Morning Bridge a good company?” is too broad for this first task. Good might mean customers like its products, its operations are profitable, its finances are resilient, or its stock is inexpensive. If you put all those questions under one label, you can collect a great deal of material without knowing which question you have answered.

A more useful starting question is: “Did second-quarter revenue growth come mainly from the original locations or from the new ones?” It identifies a business, period, and distinction that can test the initial explanation that rising totals reflect a stronger brand.

Write down two competing explanations. First, the original locations improved, with new locations adding further growth. Second, expansion increased the total even though the original locations may not have improved. They are not necessarily mutually exclusive. You do not need precise probabilities at this stage. Ask what information would help distinguish them.

The next useful evidence would be revenue broken down between the original and new locations, together with their operating periods. More photographs of a busy store would not answer the same question.

A same-store sales measure generally compares sales at a defined group of comparable locations across periods. If the company reports this measure, read its definition. Which locations qualify? Over what period? Has the calculation changed? A familiar metric name is not a guarantee that two numbers are comparable. SEC guidance uses same-store sales as an example when explaining the need for metric definitions, calculation methods, and explanations of changes. [2]

Record the limit of the available evidence: “If comparable-location and operating-period information is unavailable, I can describe total revenue growth but cannot attribute it to improvement at existing stores.” Missing information can also be a reason to end this round of investigation. Save the gap and what would let you reopen the question, rather than keep collecting material that does not answer it.

Look for evidence that could change your mind

If you enjoy Morning Bridge's products, supportive explanations may come easily. To challenge your explanation, complete this sentence: “I would become less confident in my explanation if…”

One useful answer is: “Revenue at the original ten stores fell, while the new stores supplied the increase in the total.” That result would weaken a claim that the existing locations were enjoying broad improvement. It is specific, observable, and connected to the question.

“I will reconsider if the economy gets worse” is harder to use because it does not say what evidence to check or which part of your reasoning would change.

Contradictory evidence deserves scrutiny too. A person saying, “The store was empty when I visited last week,” does not automatically overturn a quarter of recorded orders. The observation may still be informative, but record its location, time, and limits. Do not treat appealing anecdotes as strong evidence while dismissing inconvenient figures as noise.

One well-defined piece of relevant evidence can be more useful than ten repeated opinions; quality matters more than a count of positive and negative claims.

Save a first version before opening the next packet

Condense your work into six items. Here is a model note. Your wording can differ, but preserve the distinction between a conclusion and an unanswered question.

1. Question: Did second-quarter revenue growth come mainly from original locations or new locations?
2. Checked information: M1 reports revenue, orders, and ending store counts for both quarters. Revenue and orders each increased 12%; average revenue per order was $10 in both quarters.
3. Working explanation: Expansion may have contributed to growth. A stronger brand is a possible explanation, not a conclusion established by the totals.
4. Missing evidence and a challenge: Comparable revenue for the original stores and operating periods for the new stores are missing. If I had assumed demand improved at the original stores, falling revenue would prompt me to recheck that assumption.
5. Next action: Request the original/new-store revenue breakdown and its definitions. Do not value the business or decide to buy from this note alone.
6. Version: First note, based only on M1 and M2. Save later changes separately rather than overwriting this version.

This note makes your reasoning inspectable. Another reader can find the figures, see why an explanation remains open, and identify what would change your judgment. When you return to it later, you will not have to reconstruct the original reasoning from memory.

Saving a version also protects you from quietly rewriting the past. Once you learn what happened next, an outcome can start to feel obvious. The first note preserves the information you actually had, including your uncertainties. Comparing it with later evidence helps you learn how your view changed.

Pause here: Using only M1 and M2, fill in your own six-item note and save it as version one. Include an unconfirmed explanation and the evidence you would request. Finish before reading the next packet; do not backfill this version with later information.

The second packet: update your note

Now open exhibit M3. It reports that the original ten locations generated $980,000 in the second quarter, while the two new locations generated $140,000. Together, those amounts equal the reported $1,120,000. The original-store group is the same ten stores included in the first quarter. Operating-day, seasonal, promotional, and cost details are still unavailable.

Revenue at the original stores fell from $1,000,000 to $980,000, a decline of 2%. The new locations added $140,000, more than offsetting the $20,000 decline at the original group. The business therefore recorded a net increase of $120,000. In this example, growth in the total clearly does not establish growth at the original stores.

Do not jump from one unsupported conclusion to another. A decline at the original locations does not, by itself, prove that the brand is deteriorating. Revenue contributed by new locations does not prove that expansion is profitable or unprofitable. We still need to investigate costs, the reasons for the decline, and whether the pattern continues.

A second version of your note could say: “M3 shows that new locations supplied this quarter's net revenue growth, while revenue at the original locations declined slightly. The first note's possible explanation of a stronger brand remains unsupported; declining revenue alone also does not prove declining demand. Next, check operating periods, prices, and costs before judging the quality of the growth.”

Practice: classify three statements

Break each statement below into its separate claims. Decide whether each claim is supported by the materials, remains an explanation to test, or cannot currently be concluded. One sentence can contain more than one type of claim; explain your choices.

• “The company's second-quarter revenue was 12% higher than first-quarter revenue.”
• “The brand is stronger, so profit will definitely increase in the next two quarters.”
• “Revenue at the original stores declined 2%, so the new stores must be losing money.”

The first statement can be checked using M1, provided you retain the period and revenue definition.

In the second statement, a stronger brand remains an explanation to test. A definite increase in future profits cannot currently be concluded: future revenue and costs are among the missing links.

In the third, the first half is supported by M3, but the conclusion about new-store profits is not: no new-store cost information was supplied.

For an application exercise, suppose you can request only one more item. You want to assess whether performance at the original locations actually deteriorated. Which would you request first?

• Operating days and comparable-store definitions.
• An optimistic interview with management.

The first is more directly useful. M3 already supplies the revenue breakdown; the next task is to check operating time and comparability. Those details alone would still not establish whether profits deteriorated.

An interview can suggest questions, but it cannot replace operating-day data and comparable-store definitions. If your question instead concerned how management explains its expansion policy, the interview would have a different role.

Try a different business

A second fictional business, a bakery, reports revenue rising from $800,000 to $880,000 across two quarters using the same revenue definition. It has eight locations at each quarter-end, with no cost or operating-day information. Write three sentences: what is supported, what remains unknown, and what you would check next. Write your answer before reading the guidance.

Your takeaway: a one-page research note

Keep the first note you saved earlier alongside the version updated after M3. Include:

• One item attributed to M1, M2, or M3, alongside your interpretation.
• An unresolved question and evidence that would prompt you to reconsider.
• A next research action and an explanation of what changed in the second version.

Your note should distinguish the 12% increase in total revenue from the 2% decline at original locations. If you wrote version one after seeing M3, label it a retrospective reconstruction, not a prior judgment. Add the three-sentence assessment from the new situation above to check whether you can apply the method elsewhere.

In brief: turn an impression into a researchable judgment

• Define the question before collecting evidence. “Is it a good company?” mixes business quality, profit, and price. Asking where growth came from identifies the breakdown you need.
• Separate a measured result from its cause. The 12% increase is a result; a stronger brand is an explanation. Store-level information can change that explanation without answering the profit question.
• Update the reasoning, not just the opinion. Keep the earlier note and identify which step new evidence changes. When a necessary input is missing, narrow the conclusion rather than fill the gap with a story.

Next: G12: facts, interpretations, and forecasts develops the process of checking sources.

References

1. Investor.gov: Using EDGAR to Research Investments. Supports the discussion of locating company filings. The one-page note and Morning Bridge example are original teaching designs, not SEC-prescribed templates.

2. SEC: MD&A Key Performance Indicator Guidance, 2020, pages 3–5.

Manuscript status

• AI-assisted manuscript. Human financial review and target-reader testing are pending. This draft has not been publicly released.
• Shared synthetic case materials and calculation notes