How to fill in a Business Model Canvas: step-by-step with real examples

The Business Model Canvas looks simple. Nine blocks. One page. You can explain it in five minutes. But filling it in well is a different story.

I have facilitated over 100 Business Model Canvas sessions in 15 years. With manufacturing companies, industrial equipment makers, technology firms, and B2B service companies. The pattern is consistent: most teams fill in the canvas the wrong way. Not because they are bad at strategy. Because nobody showed them the right order, the right questions, or the common traps per block.

Most guides walk you through the nine blocks left to right, top to bottom. That is the layout order, not the thinking order. The result: teams start with Key Partners or Key Activities, spend most of their energy on what they already know, and run out of time before they get to the blocks that actually determine whether the business model works.

This article gives you the order that works in practice, the questions most teams forget to ask, and the specific traps I see teams fall into per block.

Before you start: who should be in the room

The biggest factor in how good your canvas turns out is not the facilitator, the template, or the sticky notes. It is who fills it in.

Get three to five people from different functions. At minimum: someone close to customers (sales, account management), someone who understands operations, and someone responsible for the business. If those three perspectives are not in the room, your canvas will have blind spots.

I once facilitated a session at an industrial components company where the CEO filled in “direct sales” under Channels. The sales director, sitting across the table, said: “We have not done direct sales in three years. Everything goes through distributors now.” That single correction changed four blocks on the canvas.

The disagreements are the point. When the head of sales sees the business model differently from the product manager, you have found something worth exploring.

One more thing: time-box each block to 8-10 minutes. Without a time limit, teams spend 40 minutes on Customer Segments and rush through Cost Structure in 5 minutes. Every block deserves the same disciplined attention.

The order that actually works

Here is the sequence I use in my sessions. It is not the layout order. It starts with the customer and builds outward.

The logic: blocks 1-5 define the value side of your business model (the customer-facing half). Blocks 6-9 define the infrastructure side (the operational half). When you fill in value first, you design the operations to deliver that value. When you start with operations, you design the value around what you already have. That is the difference between a business model designed for the customer and one designed for your comfort.

Step 1: Customer Segments

The key question: Who specifically are you creating value for?

This is where most canvases go wrong before they even start. Teams write “SMEs” or “manufacturing companies” or, the worst one, “everyone.” Those are categories, not segments.

A useful customer segment describes a group of people with the same job to be done, the same pains, and the same willingness to pay. “Head of quality at automotive tier-1 suppliers with 500+ employees, responsible for passing OEM audits” is a segment. “Manufacturers” is not.

The trap: Listing too many segments on one canvas. If you serve three genuinely different customer segments, you have three business models. Mixing them on one canvas produces a blurred picture where nothing is specific enough to act on. Fill in a separate canvas for each segment, or at minimum, pick the one segment you want to design for now.

Real example: A specialty chemicals company I worked with had “chemical industry” as their customer segment. When we dug deeper, they actually served three very different buyers: plant managers at bulk chemical producers (buying on price and reliability), R&D directors at specialty formulators (buying on technical performance), and procurement officers at multinationals (buying on compliance and supply chain risk). Each had different jobs, pains, and gains. Each needed a different value proposition. One canvas could never capture all three.

Step 2: Value Propositions

The key question: What problem are you solving or what gain are you creating for this specific customer segment?

Notice the words “for this specific customer segment.” Your value proposition is not a list of features. It is the match between what you offer and what the customer actually needs.

Write outcomes, not features. Not “AI-powered analytics platform” but “know which production line will fail before it fails, reducing unplanned downtime by 40%.” Not “modular design” but “reconfigure your production line in hours instead of weeks.”

The trap: Writing your value proposition from the inside out. Teams describe what they built, not what the customer gets. The test is simple: read your sticky note out loud and ask “so what?” If the answer to “so what?” is the real value, you wrote a feature, not a value proposition.

Real example: An equipment manufacturer had “German engineering quality” as their value proposition for years. In a session, I asked their three biggest customers what they actually valued. The answer was not quality. It was that the equipment ran for 18 months between service calls. The quality was a feature. The 18 months of uninterrupted production was the value. That distinction changed their entire sales approach.

If you struggle with this block, stop and do a Value Proposition Canvas session first. It zooms in specifically on the fit between your customer’s world and your offering. For a step-by-step guide to the value proposition side, see How to fill in a Value Proposition Canvas.

Step 3: Channels

The key question: Through which channels does your customer segment want to be reached, and how do you deliver your value proposition to them?

Channels are not just marketing channels. The block covers the full customer journey: how customers discover you, how they evaluate your offer, how they buy, how you deliver, and how you provide after-sales support.

The trap: Listing “website” and “social media” because they sound modern. Channels must match the customer segment. If you sell complex industrial equipment to plant directors, your primary channel is probably direct sales with technical demonstrations, not Instagram. If you sell standardized components to procurement departments, a digital ordering platform makes sense. The channel follows the customer, not the other way around.

Real example: A B2B SaaS company targeting manufacturing companies listed “content marketing” as their primary acquisition channel. Their average deal size was €85.000 per year with a 9-month sales cycle involving 4 stakeholders. Content marketing was part of the awareness phase, but the actual channel was direct sales supported by technical proof-of-concepts. Once they mapped the full channel honestly, they reallocated 60% of their marketing budget.

Step 4: Customer Relationships

The key question: What type of relationship does each customer segment expect, and what does it cost you to maintain it?

This block is often confused with Channels. Channels are how you reach customers. Customer Relationships are what the ongoing interaction looks like. Self-service? Personal assistance? Dedicated account manager? Community? Co-creation?

The trap: Writing what you wish the relationship was instead of what the customer expects. A startup might want a self-service, low-touch model because it scales. But if your customer segment is enterprise procurement, they expect a named account manager, quarterly business reviews, and a direct phone number. Designing for the relationship you want instead of the one the customer requires is how you lose deals.

Real example: A building materials manufacturer wanted to shift from “relationship-based sales” to a digital self-service portal. Good idea in theory. The problem: their customers, construction project managers, relied on the sales reps to configure complex orders and check product compatibility. Removing that relationship meant the customer had to do the engineering work themselves. The digital portal launched and adoption was 8%. The relationship was not a cost to cut. It was part of the value proposition.

Step 5: Revenue Streams

The key question: For what value are customers actually willing to pay, and how do they prefer to pay?

Note: “for what value,” not “for what product.” Customers pay for outcomes. The revenue model should reflect what the customer values, not just what you sell.

The trap: Only listing your current pricing model. This block is about revenue logic, not just a price tag. Ask: Could this be a subscription instead of a one-time sale? A usage-based fee instead of a license? A performance-based model where you share in the customer’s gain? Even if you do not change your model today, exploring alternatives reveals assumptions about how your customers define value.

Real example: A sensor manufacturer sold hardware for €2.500 per unit. When they mapped the Value Proposition block honestly, they realized their customers did not value the sensor. They valued the data the sensor produced, specifically the predictive maintenance alerts that prevented €50.000 production shutdowns. That insight led them to test a sensor-as-a-service model: free hardware, €400/month for the data service. Their revenue per customer tripled.

Step 6: Key Resources

The key question: What assets are absolutely required to make this business model work?

Key Resources come in four types: physical (factories, machines, logistics), intellectual (patents, brand, proprietary data), human (specialists, sales team), and financial (credit lines, cash reserves).

The trap: Listing everything you have instead of what you need. Your office building is not a key resource unless it is genuinely part of your value delivery. A key resource is something that, if you lost it tomorrow, would break the business model. For a pharmaceutical company, that might be a patent portfolio. For a consulting firm, it is the expertise of specific people. For a manufacturing company, it might be a specialized production line that competitors cannot replicate.

Real example: A food processing equipment company listed 15 resources on their canvas. When I asked “which three, if gone, would kill the business model?”, the room got quiet. The answer: their application engineering team (customers buy because of the technical consultation, not the machine), their installed base data (which drives their service revenue), and their regulatory certifications. Everything else was operational support, not a key resource.

Step 7: Key Activities

The key question: What must you do exceptionally well to deliver your value proposition through your channels?

The trap: Listing daily operations instead of strategic activities. “Paying invoices” is not a key activity. “Developing custom formulations for each customer segment” is. Key activities are the things that, if you stopped doing them, would break the connection between your value proposition and your customer.

Real example: A packaging company’s canvas listed “production” as their key activity. That was true but not useful. When we pushed further: their real key activity was co-development, working with customers 6 months before launch to design packaging that met technical specs, regulatory requirements, and shelf appeal simultaneously. Production was the output. Co-development was the activity that won deals.

Step 8: Key Partners

The key question: Who do you need to make this business model work, and what do they provide that you cannot or should not provide yourself?

The trap: Listing every supplier and vendor as a key partner. Your office cleaning company is a supplier, not a key partner. A key partner provides something that is part of your value delivery. A distribution partner who owns the customer relationship. A technology partner whose platform your product runs on. A certification body whose approval your customers require.

Real example: An industrial IoT company realized through this block that their real key partner was not their cloud provider (easily replaceable) but the machine OEMs who gave them access to install sensors on their equipment. Without that partnership, they had no data, no product, and no business model. That realization changed their entire partnership strategy from technology partnerships to OEM integration agreements.

Step 9: Cost Structure

The key question: What are the most significant costs in operating this business model, and which key resources and key activities drive those costs?

This block comes last for a reason. Only now can you look at the full picture and ask: does the revenue from block 5 exceed the costs required by blocks 6, 7, and 8?

The trap: Being vague. “Salaries” is not helpful. “12-person application engineering team at an average loaded cost of €95.000 per person” tells you something you can work with. The more specific your cost structure, the better you can identify where the model might break.

Real example: A B2B platform company discovered, through this exercise, that their customer acquisition cost was €12.000 per customer (sales team, demos, proof-of-concepts) while their average first-year revenue per customer was €8.000. The business model was structurally unprofitable until year two. That was not a problem, but it was an assumption that needed to be visible and tested. Without mapping it explicitly, they would have kept burning cash wondering why growth was not translating into profit.

After you fill it in: three things to do before you leave the room

Filling in all nine blocks is not the finish line. It is the starting point. Here is what separates teams that get value from the canvas from teams that produce expensive wallpaper.

1. Mark your confidence levels. Go through every sticky note and mark it green (you have evidence), yellow (you have some data), or red (you are guessing). First-time canvases are mostly red and yellow. That is normal. The problem is not having assumptions. The problem is not knowing which ones are assumptions.

2. Identify the three riskiest assumptions. Look at the red sticky notes. Which ones, if wrong, would sink the entire business model? Those are your priority. Not the easiest to test. The most dangerous.

3. Plan your first tests. For each high-risk assumption, design a simple test you can run in two weeks. A customer interview. A landing page. A pricing conversation. This is where the Business Model Canvas connects to testing business ideas. The canvas generates hypotheses. Testing generates evidence. One without the other is incomplete. For a complete process on testing those assumptions, see How to test business assumptions.

If you want to go deeper into how the canvas fits into a systematic innovation process, read about innovation portfolio management, which shows how to manage multiple business model experiments across a portfolio. And if you are not sure whether your organization is set up to do this kind of work, start with an innovation readiness assessment.

For a deep dive into each of the nine blocks, see the 9 building blocks of the Business Model Canvas.

Want to see completed canvases from real companies? Read Business Model Canvas examples for B2B, manufacturing, and service business models.

Once your canvas is filled in, the next step is Business Model Canvas validation: turning your assumptions into tested knowledge.

This step-by-step approach is part of a broader discipline called business model innovation: the systematic practice of designing, testing, and improving business models.

Quick reference: questions and traps per block

The Business Model Canvas was designed by Alexander Osterwalder as a thinking tool, not a planning tool. Filling it in is not the goal. Using it to surface assumptions, spark disagreements, and identify what needs testing, that is the goal.

If you compare the Business Model Canvas with the Lean Canvas, the biggest difference is exactly this: the BMC maps the full business model system, while the Lean Canvas focuses on problem-solution fit. Choose based on what stage your project is in.

Frequently asked questions

What order should you fill in the Business Model Canvas?

Start with Customer Segments, then Value Propositions, then Channels. This customer-first sequence forces you to design the business model around a real need instead of your product. Only after you understand who you serve, what they value, and how you reach them should you move to the infrastructure blocks: Key Resources, Key Activities, Key Partners. Finish with Revenue Streams and Cost Structure to check whether the model is financially viable.

How long does it take to fill in a Business Model Canvas?

A first canvas takes 60 to 90 minutes with the right team in the room. But the first version is never the final version. Plan for at least two iterations: one to get your assumptions on paper, and a second after initial customer conversations to update what you learned. Teams that rush through in 30 minutes produce generic canvases that help no one.

Can one company have multiple Business Model Canvases?

Yes, and most companies should. If you serve different customer segments with different value propositions through different channels, each combination is a separate business model. A manufacturing company selling standard products through distributors and custom solutions through direct sales has two business models. Mapping them on separate canvases makes the differences visible and prevents the common mistake of blending everything into one generic canvas.

What is the difference between filling in a Business Model Canvas and a Value Proposition Canvas?

The Business Model Canvas maps your entire business model across 9 blocks. The Value Proposition Canvas zooms in on the fit between one customer segment and one value proposition. Use the Value Proposition Canvas first to deeply understand your customer’s jobs, pains, and gains. Then use the Business Model Canvas to design the full business model around that validated value proposition.

How many sticky notes should each block have?

Aim for 3 to 5 sticky notes per block. Fewer than 3 usually means you have not thought deeply enough. More than 7 means you are listing everything instead of making choices. Each sticky note should contain one idea, written in a short phrase that anyone in the room can understand without explanation. If a sticky note needs a paragraph to explain, break it into multiple notes or simplify the language.