What a Business Model Actually Is
A business model describes how an organisation creates value for customers and captures a portion of that value as revenue and profit. The confusion between business model and business plan, or between business model and revenue model, is worth clarifying: a business plan is a document; a business model is a system. A revenue model is one component of the business model — how the business charges for what it provides. The business model is the complete answer to: who are the customers, what problem are we solving for them, how do we deliver the solution, and how do we make money doing it?
The business model question that most business planning skips: why would the business be able to earn a profit sustainably rather than having that profit competed away by rivals offering the same thing? The business model that can be copied immediately by any well-capitalised competitor isn’t a business model — it’s a temporary advantage. The durable business model has some combination of switching costs, network effects, proprietary technology, regulatory advantage, or brand that creates a competitive moat preventing profit erosion.
The Business Model Canvas: A Tool for Thinking, Not Just Documenting
The Business Model Canvas, developed by Alexander Osterwalder, provides a one-page framework for mapping the nine core elements of any business model: customer segments (who are we serving?), value propositions (what are we offering them?), channels (how do we reach them?), customer relationships (what kind of relationship do we maintain?), revenue streams (how do we earn from each segment?), key resources (what do we need to deliver the proposition?), key activities (what must we do?), key partnerships (who helps us?), and cost structure (what are the major costs?).
The Canvas is most useful not as a document to complete once but as a tool for testing assumptions: each block contains assumptions that may or may not be correct, and the exercise of making those assumptions explicit creates the opportunity to identify which are most critical to test. The customer segment assumption is only correct if customers with those characteristics actually exist in sufficient numbers to support the business. The value proposition assumption is only correct if customers value what the business delivers enough to pay the proposed price. The Canvas makes these testable hypotheses visible.
Subscription vs. Transaction: The Revenue Model Decision That Defines Everything
The shift from transaction-based revenue (pay once for the product or service) to subscription-based revenue (pay regularly for ongoing access) has been one of the most significant business model developments of the past two decades. Subscription revenue is more predictable, more valuable per dollar (the market typically values recurring revenue at higher multiples than one-time revenue), and more defensible (the switching cost of cancelling a subscription is real, even if the cost is just inertia).
The subscription model decision isn’t appropriate for every business — physical products with long replacement cycles, bespoke services, and one-time projects don’t lend themselves naturally to subscription structures. The businesses that have successfully moved to subscription models from transaction models (Adobe’s shift from software licences to Creative Cloud, John Deere’s shift toward maintenance subscriptions) have done so by identifying the ongoing value that the subscription funds, not just by relabelling a one-time purchase as a recurring charge.
Unit Economics: The Business Model Test That Reveals Viability
Unit economics — the revenue and cost associated with a single customer, transaction, or unit of product — reveal whether the business model is viable before scale demonstrates it expensively. A business with negative unit economics (where acquiring and serving one customer costs more than that customer generates) cannot be scaled into profitability; the losses scale alongside the revenue. Understanding unit economics early enough to change the business model is the difference between a pivot and a failure.
The unit economics calculation for most businesses: Customer Acquisition Cost (the total marketing and sales spend divided by the number of new customers it produces in a period) compared against Customer Lifetime Value (the average revenue per customer times the average number of purchases times the gross margin, for the duration of the average customer relationship). When LTV exceeds CAC by at least 3:1, the business model is generally considered viable for scaling. When the ratio is below 3:1, either the revenue side needs to increase, the acquisition cost needs to decrease, or the model needs rethinking.
Evolving the Business Model Without Destroying the Business
Business model evolution — changing core elements of how the business creates and captures value — is necessary for most businesses over a long enough time horizon, as customer needs change, competition intensifies, or technology creates new possibilities. The difficulty is that business model change is inherently disruptive: existing customers, employees, and partners have expectations built around the current model, and changes that improve the model in the long term often create disruption in the short term.
The business model evolution approach that produces the best outcomes: run the new model as an experiment alongside the existing model before committing to the transition. A software company exploring a subscription model can offer subscriptions to new customers while maintaining licence sales to existing ones; the experimental track produces revenue data, customer feedback, and operational learning before the full transition is committed to. This parallel operation approach is more expensive than a clean cutover but much less risky, particularly when the existing model is currently profitable.