Lin likes a store, shops there often, and recommends it to friends. Her research note says, “Loyal customers and a deep moat.” A friend asks who pays for the low prices. Lin returns to a receipt: customers paid the company, but how much of that money could the business keep?
Follow a customer's payment before assigning labels such as great product, strong brand, or competitive advantage. A business explanation needs to connect revenue with the costs of delivering it. Then an attractive description can become a research claim you can check.
What you will be able to do
- Describe customers, needs, products, payment arrangements, and costs in one business model.
- Separate volume, price, and product mix, and explain how fixed costs amplify changes in profit.
- Identify evidence and counterevidence for an advantage and save a business map with observable review measures.
Who uses the product, and who pays?
“Retail,” “software,” and “health care” identify broad activities without describing how revenue arrives. Software can be sold through one-time licenses, recurring subscriptions, usage charges, or advertising that lets users participate without paying. An industry label does not choose among those arrangements.
Write down user and payer separately. What need is being addressed, what alternatives exist, and how often is the payment decision renewed? If users and payers differ, user growth alone can miss the person deciding the revenue budget.
A business model connects whom the company serves, what it provides, how it delivers, how it charges, and what it must spend. CFA Institute's public overview places customers, suppliers, operations, and risks within business-model analysis.[1] The map below is a way to inspect those connections, not an automatic quality score.
Follow the contract too. An order, delivery, revenue recognition, and collection may happen on different dates. Label those events separately even in a simple diagram. Treating the order total as both revenue and cash received would make the picture simpler by removing an important question.
A useful map can stop at a missing link. If the payer is clear but collection terms are not, mark those terms as unresolved. The diagram should show the limit of current understanding rather than supply contractual details based on what similar companies usually do.
Begin with a fixed historical business description
For C1-R supplemental retrospective reading, use Costco's fiscal 2025 10-K, covering the year ended August 31, 2025. It is later historical learning material, outside the original frozen C1 information packets, not a current operating or price assessment. Item 1 and Item 7 describe merchandise sales, membership fees, and membership behavior.[2]
Separate the payment for membership access from payments for goods. The research questions then include membership counts and charges, shopping frequency, purchase size, and delivery costs. Do not multiply a period-end membership count by a current annual fee and call it full-year recognized revenue. Enrollment timing, categories, geography, and recognition periods must be checked.
Renewal can offer evidence about repeated choice, but its definition, population, and observation window matter. The cost of providing the service still matters as well. Member satisfaction and a satisfactory shareholder return remain separated by operating costs, investment requirements, and the price paid for the stock.
The point is to find the issuer's actual definitions and distinguish them from your estimates. Do not insert figures from the fictional stores below into Costco's map simply because both involve retail customers.
Rebuild revenue from volume and price
Now use Pine Brook, an independent fictional neighborhood store. All figures are synthetic US dollars and do not describe Costco or another real business. Ignore returns, taxes, inventory timing, and other activities for this example.
In the first period, 100 units sell for $100 each, generating $10,000. Selling 110 units at the unchanged price produces $11,000: 10% growth driven by volume. Keeping volume at 100 but charging $110 also produces $11,000 and 10% growth, through a different mechanism.
If both volume and price increase to 110, revenue becomes $12,100, a 21% increase. Adding 10% and 10% misses the higher price on the ten additional units. A decomposition needs to rebuild the final total, not merely attach plausible growth labels.
Why distinguish mechanisms when the first two totals match? More units may require additional supply, staff time, or capacity. A higher price raises questions about willingness to pay and whether customers switch. Equal revenue growth does not imply equal operating conditions or equal future durability.
A higher average price can happen without a price increase
Consider a separate two-product example. One product costs $100 and the other $200. Selling 50 units of each generates $15,000 across 100 units, or $150 per unit on average.
Next period, 40 units of the cheaper product and 60 of the expensive one sell. Total volume remains 100, but revenue becomes $16,000 and the average becomes $160. Neither product's price changed. Customers bought a different mix.
Calling that average increase “pricing power” would answer the wrong question. Mix research asks whether the preference for the higher-priced item can persist. A same-product increase asks whether customers accept a higher charge for a comparable purchase. The evidence and alternatives differ.
Break volume and price into useful product or geographic groups where the source allows. Check that groups are complete and nonoverlapping before aggregating them. If only total revenue is available, record the driver as unknown. An AI-generated breakdown based on general industry knowledge is not a disclosed company table.
The mix example also supplies a quick check for a summary. Ask it to identify which individual price changed. In these inputs, neither did. If the explanation still calls the increase a same-product price rise, return it to the component quantities. A mathematically correct average cannot rescue a mistaken account of its cause.
An average remains useful when its construction is understood. The error is allowing its familiar label to erase the components. Keep the group prices and quantities beside the average so another reader can see which change produced it.
Why a 10% sales decline can cut profit 20%
Give Pine Brook another independent, simplified cost model. It sells 100 units at $100, with a variable cost of $60 per unit and $2,000 of fixed operating costs for the period. These are all the supplied operating costs; interest and taxes are excluded.
Revenue is $10,000 and variable costs are $6,000. Subtract the $2,000 fixed cost and simplified operating profit is $2,000. Variable costs move with the unit count in this model. Fixed costs remain unchanged only within the stated period and operating range. Actual leases, staffing, and capacity can involve thresholds rather than remaining fixed forever.
At 110 units, with price and unit and fixed costs unchanged, revenue is $11,000 and variable costs are $6,600. Profit becomes $2,400. Ten percent more volume creates 20% more profit because the fixed cost does not rise proportionately.
At 90 units, revenue is $9,000, variable costs are $5,400, and profit falls to $1,600. Profit drops 20% although volume drops 10%. The same structure amplifies both improvement and deterioration. It cannot supply extra profit in the optimistic scenario while disappearing automatically in the adverse one.
This is the intuition behind operating leverage: a cost structure can make profit respond more strongly than sales to changing activity. The result depends on the cost assumptions. It does not show that a company with operating leverage is necessarily a good investment.
We have not calculated cash flow. Expenses and cash payments may have different dates, and expansion may require inventory, equipment, and working capital. Higher model profit does not mean the same additional dollars are immediately available for distribution to shareholders. Later financial chapters examine that bridge.
What would make an advantage difficult to copy?
An economic moat is a metaphor for a relatively durable competitive advantage. It usefully directs attention to competitors. It becomes misleading if it suggests a permanent physical wall. Technology, customer needs, and rival offerings can change the conditions supporting the advantage.
Suppose Pine Brook's purchasing scale improves buying terms. A candidate mechanism is lower unit costs that support attractive prices or services and keep customers returning. Scale can also add management complexity, and low prices can reflect sacrificed profit. Check purchasing and cost evidence instead of awarding an advantage for store count alone.
A brand claim needs customer behavior. Do repeated purchases persist without larger discounts? What happens after a service failure? Can a brand established in one region travel into another? “People like the store” starts the inquiry but does not complete those tests.
For switching costs, identify who bears which cost: training staff, moving data, or testing substitutes, for example. Such costs may support retention, yet open standards or better products can reduce them. A customer may also remain only because a short contract has not expired. Staying does not automatically establish lasting satisfaction.
These are possible mechanisms, not facts about Pine Brook or Costco. Each arrow from advantage to result needs an alternative explanation and evidence that could weaken it. Low prices with worsening cash pressure, retention dependent on subsidies, or deteriorating service after expansion should be allowed to change the original judgment.
Replace adjectives with the next observation
A business map does not need dozens of metrics. Pick two or three that most affect the current claim. For repeat purchasing, examine consistently defined retention or shopping frequency along with discounts and customer-acquisition costs. For scale savings, examine unit costs, mix, and service quality.
Missing disclosure is a gap, not a reason to create unlimited data-collection tasks. Compare the business description, risk discussion, and management explanation for evidence relevant to the question. Those passages are investigation routes, not optimistic and pessimistic votes to count.
Label reported facts, estimates, and hypotheses separately. Period revenue from a filing is a scoped report value. Quantity inferred from an average price is your estimate. A belief that customers will remain after a price increase is a hypothesis. They can appear on the same map if their different status remains visible.
Choose a review trigger that could affect an arrow rather than a broad instruction to follow the company. If the claim depends on unit-cost improvement, new cost or mix evidence matters more than another general statement about the brand. Limited attention can then follow the judgment's actual weak point.
Rebuild the business before checking the answers
Explanation question: Why can average revenue per unit increase without higher prices for the same products?
Application question: In Pine Brook's fixed-cost model, why does moving from 100 to 90 units cut profit by more than 10%? Write quantity, revenue, variable cost, and profit before reading the guide.
For the first question, more expensive products can take a greater share of the mix. Check individual prices and quantities rather than infer a price increase from the average. For the second, $9,000 revenue less $5,400 variable costs and $2,000 fixed costs leaves $1,600. The fixed cost did not decline with activity, so profit fell 20% from $2,000.
Change the cost conditions
Pine Brook now sells 110 units at $100, but variable cost rises to $65 per unit and fixed costs rise to $2,200. Nothing else changes. Calculate operating profit, then assess whether 10% more units proves a scale advantage. What evidence is still needed?
Revenue is $11,000, variable cost is $7,150, and profit after fixed costs is $1,650. That is 17.5% below the original $2,000. The earlier $2,400 answer depended on unchanged costs and cannot be carried into this case.
Investigate whether cost changes arise from purchase prices, service investment, mix, or efficiency. One profit decline also does not establish that scale can never help. The changed numbers undermine a particular inference; they do not authorize an unsupported permanent conclusion in the opposite direction.
Save an updateable business map
Use the fixed Costco filing or a public company of your choice. Map customer, need, product, channel, revenue sources, main costs, and cash events. Attach a source or estimation method to each item and mark unknowns. Keep the fictional model's numbers out of real-company fields.
Choose one advantage claim. Record its mechanism, available evidence, two alternative explanations, one countercondition, and the next metric and review date. To use C1-B Target instead, first obtain its own disclosures and structure. Membership-warehouse assumptions cannot migrate merely by changing the company name.
Cover the words strong, leading, and high quality on your map. Does it still explain who pays, why costs arise, and how profit remains? If only an introduction and “deep moat” are left, return to the arrow with the weakest evidence.
Follow the payment through every connection
Start with payer and need, distinguish volume, price, and mix, then connect revenue to costs, profit, and cash. Keep units, periods, and assumptions visible rather than let revenue growth stand in for what shareholders ultimately receive.
An advantage claim needs a mechanism competitors find difficult to copy and conditions under which that mechanism could fail. Better business understanding supports better questions. Valuation and risk still need separate work. The missing connection on the map identifies the next evidence request.
References
- CFA Institute: Business Models. Public 2026 overview; paid text not accessed.
- Costco fiscal 2025 Form 10-K. Year ended August 31, 2025; Item 1 and Item 7. Fixed C1-R reading, not current results. All Pine Brook models are separate synthetic inputs.
Sources checked October 30, 2026. AI-assisted draft; a separate AI reviewer has completed an independent full-text review; human financial review and target-reader testing remain pending.