What each of the nine blocks means
Customer Segments: the distinct groups you serve. Split a group out whenever it needs a different offer, channel or relationship — that test is what stops "everyone" from becoming a segment. Value Propositions: the job you get done or the pain you remove for each segment, described as the change the customer experiences rather than as a feature list. Channels: how people find out about you, evaluate you, buy, receive the thing, and get help afterwards. Customer Relationships: what kind of ongoing relationship each segment expects — self-serve, guided onboarding, community, a named account manager.
Revenue Streams: what customers actually pay for and in what shape — one-off sale, subscription, usage, licensing, advertising. Key Resources: the assets the model can't run without, whether people, intellectual property, capital, data or physical plant. Key Activities: the things you must do repeatedly and well for the value proposition to hold. Key Partners: who supplies, distributes or de-risks what you don't want to build yourself. Cost Structure: what running all of the above costs, split into what is fixed and what scales with volume.
The order to fill it in
Start on the right, not the left. Customer Segments first, then Value Propositions, because everything else is downstream of who you serve and what you offer them. Then Customer Relationships and Channels — how you reach and keep those people — followed by Revenue Streams, which closes the customer-facing half of the canvas.
Only then move left to the infrastructure that delivers it: Key Activities, Key Resources and Key Partners. Cost Structure comes last, because it is a consequence of the left-hand blocks rather than an independent choice. Teams that fill the canvas left to right almost always end up describing the company they already have instead of the model they are testing.
Common mistakes
Writing features into the Value Propositions block is the most frequent one — "real-time dashboard" is a feature, "finance stops rebuilding the same report every Monday" is a value proposition. A close second is a single Customer Segment labelled with a market rather than a group of people who behave the same way. If your value proposition needs two different sentences for two groups, those are two segments, and the canvas should show them.
Beyond that: Channels quietly listing aspirations rather than routes that exist, Revenue Streams stating a price with no volume assumption behind it, and Cost Structure omitting the founders' own time. The canvas is also a snapshot, not a plan — it says nothing about sequencing, funding or competition, so treat it as one artefact next to a roadmap and a market view rather than a substitute for them.
Business Model Canvas or Lean Canvas?
Use the Business Model Canvas when there is a business to describe: existing revenue, real partners, an infrastructure that costs money to run. It is strongest for mapping how an operating model fits together, comparing it against a competitor's, or working out which block a proposed change actually touches.
Use the Lean Canvas — Ash Maurya's adaptation — when the business does not exist yet and the open question is whether anyone has the problem. It swaps four blocks (Key Partners, Key Activities, Key Resources and Customer Relationships become Problem, Solution, Key Metrics and Unfair Advantage) because at that stage partners and infrastructure are guesses, while the problem and the metric that would prove it are the things worth arguing about.
Keeping the canvas where the work is
A canvas is worth more when it is revisited than when it is filled in once. Save this one to OmniCanvas and it lands on an infinite canvas as nine editable blocks, next to the customer interview notes, the pricing spreadsheet and whatever else the model depends on — so revising a block is a two-minute edit rather than reopening a file you have to go and find.