The questionWhich part of the model would sink the business if it were wrong, and how can you find out cheaply?
Two businesses can sell the same product and be entirely different businesses. One sells power drills; another rents them by the day; a third sells the holes, as a fitting service. Each has different customers, costs, cash flows and risks. The difference is the business model: the logic by which a business creates value for customers, delivers it, and captures some of it as revenue.
This briefing gives three tools for working on a business model before it is built: a vocabulary of common models, the Business Model Canvas for setting a model out on one page, and the lean start-up method for testing the riskiest parts of the model with a minimum viable product.
Section 01: What a business model is
A business model answers three questions. How does the business create value: what does it do that someone values? How does it deliver that value: through which activities, resources and channels? And how does it capture value: who pays, for what, and how much, so that the business earns more than it costs to run?
Most businesses use one of a handful of patterns, or a mix of them. Each pattern rests on a different set of conditions that must hold for it to work, and that is where its risks lie. A subscription business lives or dies by how long customers stay; a marketplace needs enough buyers and sellers at the same time; a freemium product needs enough free users to become paying ones. Choosing a model is choosing which of those conditions to bet on.
Circular business models, which are part of the IB toolkit, are business models in exactly this sense. A product service system, in which the customer pays to use a product that the business keeps owning, captures value through use rather than sale, and turns long-lasting products from a threat to repeat sales into a source of profit.
| Model | How it earns revenue | What has to be true |
|---|---|---|
| Product sale | Customers pay once for each item | Enough margin on each sale to cover the fixed costs, and a steady flow of new or returning customers |
| Subscription | Customers pay regularly until they cancel | Customers stay long enough to repay what it cost to win them |
| Marketplace | A commission on sales between buyers and sellers | Enough buyers and sellers at the same time, each attracting the other |
| Freemium | A free basic version, with some users paying for more | Enough free users become paying ones to cover the cost of serving everyone |
| Consumables | A durable item sold cheaply, with the profit on refills or parts | Customers keep buying the consumable from you rather than a rival |
| Product as a service | Customers pay to use a product the business still owns | Products last and come back, and the business can fund the stock up front |
| Advertising | Advertisers pay to reach the users | A large, engaged audience that advertisers value |
Section 02: The Business Model Canvas
The Business Model Canvas, devised by Alexander Osterwalder with Yves Pigneur, sets a business model out on one page in nine blocks. The blocks on the right describe value for the customer: who the customers are, what the business offers them, how it reaches them, how it keeps them and what they pay. The blocks on the left describe how the business delivers that value efficiently: what it must have, what it must do, who it depends on and what it all costs.
Its strength is that it shows the model as a system. Change the customer segment and the channels, the relationships and the price may all have to change; add a delivery service and the key activities, resources and costs move with it. Because it fits on one page, it is quick to redraw, so a founder can sketch three alternative models in an afternoon and compare them.
It is best filled from the customer outwards, in short statements, each of which could be tested. Many teams write each entry on a sticky note so it can move, and mark which entries are assumptions and which are backed by evidence: that is what turns the canvas from a description into a plan of what to test.
- Customer segments
The groups the business serves. This is where the guide's segmentation and targeting (4.2) come in.
- Value propositions
What the business does for each segment that it values. The USP and differentiation (4.2) belong here.
- Channels
How customers find out, buy and receive. Promotion and place in the marketing mix (4.5).
- Customer relationships
How customers are won, kept and grown. For a service, the people and processes of the extended marketing mix (4.5).
- Revenue streams
What customers pay for, how and how much: the revenue streams of 3.3.
- Key resources, activities and partners
What the business must have, what it must do well and who it depends on. Operations (unit 5) lives here.
- Cost structure
The costs the model creates, fixed and variable (3.3), and what drives each.
Section 03: The lean start-up: build, measure, learn
The lean start-up is a method for building a new business under extreme uncertainty, associated with Eric Ries and building on Steve Blank's work on customer development. It takes its name from lean production, the approach to cutting waste that the IB course teaches at HL (5.3): the waste it cuts is time and money spent building something customers do not want.
Its core is a loop. Turn the riskiest assumption into a hypothesis that could be shown false; build the smallest thing that will test it; measure what real customers do; learn whether the hypothesis held; and decide whether to persevere with the model or pivot, changing one part of it while keeping what has been learned. The faster a team goes round the loop, the less it spends finding out what does not work.
What counts as learning is evidence about behaviour, measured against a success criterion set in advance. Sign-ups that turn into paying customers, customers who return, and customers who recommend the product are learning; total visitors, downloads and likes, which rise whether or not the model works, are what start-up writers call vanity metrics.
- Hypothesis
An assumption written so that it could be shown false, such as "at least 30 of 40 pilot households will still be paying after three months".
- Experiment
The cheapest test that could prove the hypothesis wrong, with the measure and the pass mark chosen before it starts.
- Pivot
A change to one part of the model, such as the customer segment, the channel or the revenue model, based on what the experiments showed.
- Persevere
Keep the model and test the next riskiest assumption.
Section 04: The minimum viable product
A minimum viable product, or MVP, is the smallest thing a team can put in front of real customers to test its riskiest assumption with real behaviour. It is an experiment, not a cheap first version of the final product. Sometimes it is not a product at all: a landing page that takes pre-orders, a service delivered by hand to ten customers, or a pilot run from borrowed equipment.
The right MVP depends on the assumption. To test whether people want the offer, a landing page may be enough. To test whether they keep paying, nothing short of a pilot over several months will do. To test whether the service can be delivered at the planned cost, the founders must actually deliver it, at some volume, and count the costs. Choose the MVP that tests the assumption at the top of the risk list, not the one that is most fun to build.
| MVP | What it tests | What to watch |
|---|---|---|
| Landing page | Whether people sign up or pay for an offer as described | Interest is not use. Say clearly if the product is not yet available, and refund anything taken. |
| Concierge | Whether the service solves the problem, delivered by hand to a few customers | It shows nothing about whether the service can be delivered at scale or at the planned cost. |
| Behind-the-scenes manual | Whether customers use a service that looks automated but is run by people | Customers must not be misled about how their information is handled. |
| Pre-sales | Whether customers will pay before the product exists | It creates an obligation to deliver, and a failed campaign is public. |
| Pilot | Whether a small group keeps using and paying over time | Samples are small, and early customers are keener than the mass market. |
Section 05: Common mistakes
A business model is the same as a business plan.
A business model is the logic of how the business creates and captures value. A business plan is a document that sets out the model together with the market, operations, people and finances, usually to persuade someone to fund it. A canvas takes an hour; a plan takes weeks.
A minimum viable product is the first version of the product, built cheaply.
It is an experiment to test the riskiest assumption with real customer behaviour. Sometimes it is not a product at all, but a landing page, a pre-order or a service delivered by hand.
A pivot means the founders failed.
A pivot keeps what has been learned and changes one part of the model. Persevering with a model the evidence contradicts is the real failure.
Section 06: Key terms
Start-up terms the IB course does not use are defined here, as enrichment. Terms from the course link to the platform's glossary.
- Business model Source: Enrichment
The logic by which a business creates value for customers, delivers it and captures some of it as revenue.
- Business Model Canvas Source: Enrichment
A one-page template, devised by Alexander Osterwalder with Yves Pigneur, that sets out a business model in nine blocks.
- Lean start-up Source: Enrichment
A method for building a new business by testing its riskiest assumptions in quick, cheap experiments with real customers.
- Minimum viable product (MVP) Source: Enrichment
The smallest thing that can be put in front of real customers to test an assumption with their behaviour.
- Pivot Source: Enrichment
A change to one part of a business model, based on evidence, that keeps what has been learned.
- Vanity metric Source: Enrichment
A number that rises whether or not the business model works, such as total visitors or downloads.
- Freemium Source: Enrichment
A model in which a basic version is free and customers pay for more features or capacity.
Section 07: Check your understanding
The platform's own questions, not IB questions. Try each one before opening the answer.
-
What three things does a business model explain?
Answer
How the business creates value for customers, how it delivers that value, and how it captures some of it as revenue.
-
Which blocks of the canvas describe value for the customer, and which describe the efficiency of the business?
Answer
The right-hand blocks (customer segments, value propositions, channels, customer relationships and revenue streams) describe value for the customer. The left-hand blocks (key resources, key activities, key partners and the cost structure) describe how the business delivers it.
-
Refill Round's riskiest assumption is that subscribers stay about a year. Design an MVP to test it.
Answer
A paid pilot of a few months with a small group of households, served by hand, with a success criterion set in advance, such as at least 30 of 40 households still paying after three months.
-
Give two limitations of the Business Model Canvas.
Answer
It leaves out the competition and the founders; it says nothing about when cash comes in and goes out or how the start is funded; and it ignores the costs a business imposes on others unless they are added.
-
Why does a landing-page test raise an ethical question, and how can a founder handle it?
Answer
Customers may believe they are buying a product that does not yet exist. Saying clearly that it is not yet available, and refunding any money taken, keeps the test honest.
Section 08: Use it
Put the briefing to work on Refill Round's example, then on your own idea.
- Business Model Canvas builder Build a Business Model Canvas in nine blocks, mark each entry as an assumption or as evidence, rate its risk, and get the riskiest assumptions to test first.
Section 09: What is IB content here, and what is not
Each idea in this briefing, checked against the IB Business management guide (first assessment 2024).
| Idea | Status | In the guide |
|---|---|---|
| Business models and the Business Model Canvas | Source: Enrichment | Not in the guide. Related to: Not in the guide. Its blocks draw on revenue streams and costs (3.3) and on segmentation, targeting and the USP (4.2). |
| Circular business models, including the product service system | IB toolkit | Toolkit, guide p. 14 Five models in the toolkit, suggested for units 1 and 5; each is a business model in the canvas's sense. |
| Revenue streams and the types of cost | IB content | Guide, p. 28: Types of cost and total revenue and revenue streams, both AO2, on pp. 28 to 29. |
| The lean start-up, the build, measure, learn loop and the pivot | Source: Enrichment | Not in the guide. Related to: A start-up method that uses market research methods as experiments. |
| The minimum viable product | Source: Enrichment | Not in the guide. A start-up term the guide does not use. |
Written in the platform's own words. Checked against: Business management guide, first assessment 2024 (published February 2022, updated May 2022), pp. 13 to 14, 28 to 29, 31 to 32, 34. Every business named is fictional and every figure simulated. Updated 3 November 2026.