Startup business models: compare the economics before choosing the label.
A business model connects customer value, revenue logic, delivery system and cost structure. The useful question is not which model sounds modern, but which model can create, deliver and capture value for a specific buyer under real operating constraints.
Nine common structures, each with different economic pressures
These categories are useful starting points, not rigid boxes. Real companies often combine them, and the same company can use different revenue mechanisms for different products or customer segments.
SaaS
Software access delivered as an ongoing service. Retention, cost-to-serve, multitenancy and GTM fit shape the model.
Study SaaS economics →Subscription
Customers pay repeatedly for continued access to a product, service, content or bundle.
Study subscription economics →Usage-Based
Revenue varies with measurable consumption, requiring careful value-metric and cost alignment.
Study usage economics →Platform
Creates value by enabling interactions among participant groups and can monetize through several mechanisms.
Study platform economics →Freemium
A permanent free tier supports adoption while paid tiers monetize users requiring more value.
Study freemium economics →Marketplace
Connects buyers and sellers; liquidity, trust, take rate and disintermediation are central.
Study marketplace economics →Licensing
Customers pay for permission to use technology, intellectual property or another protected asset.
Study licensing economics →Advertising
Advertisers fund access to audience attention; inventory, yield, audience quality and ad load shape economics.
Study advertising economics →Transaction-Fee
Revenue is triggered by completed events, with ticket size, volume and transaction-linked costs determining contribution.
Study transaction-fee economics →Business-model comparison matrix
Use this matrix to compare the mechanism behind each model rather than treating the model name as a strategy.
| Model | Primary revenue logic | Economic strength to watch | Typical constraint | GTM implication |
|---|---|---|---|---|
| SaaS | Recurring software access | Retention, gross margin, expansion | Acquisition cost, churn, support burden | Self-serve, product-led or sales-led depending on complexity and contract value |
| Marketplace | Transaction fee, take rate, listing or service revenue | Liquidity and repeat exchange | Chicken-and-egg supply/demand, trust, disintermediation | Often requires separate acquisition strategies for each side |
| Subscription | Recurring access or delivery | Renewal value and retention | Subscription fatigue, churn, fulfilment or content cost | Requires a repeatable reason to stay, not only a strong first purchase |
| Freemium | Free acquisition plus paid conversion | Efficient free-to-paid path | High free-user cost or weak upgrade trigger | Product experience becomes a core acquisition and conversion channel |
| Usage-based | Metered consumption | Value-price alignment and expansion with usage | Revenue volatility, bill anxiety, cost pass-through | Sales and success teams must explain usage economics clearly |
| Licensing | Right to use an asset or technology | Scalability of rights without equivalent delivery cost | Negotiation complexity, enforcement, dependency on contract scope | Often relationship-led, enterprise or partner-driven |
| Platform | Participation, access, transaction, advertising or hybrid | Network effects and interaction quality | Chicken-and-egg formation, governance and multi-side incentives | Often requires separate acquisition and subsidy logic across participant groups |
| Transaction-Fee | Fee on completed event | Transaction frequency, ticket size and retained contribution | Volume cycles, processing costs and bypass incentives | Acquisition must create enough recurring transaction activity to support event-based monetization |
Four tests before changing the model
A model change is useful only when it improves a real economic or customer problem. Start with the decision, then test the connected mechanisms.
1. Customer value
What outcome is the buyer paying for, and how frequently does that value recur?
2. Revenue logic
What event triggers payment, and does the pricing unit track the value customers perceive?
3. Cost structure
Which costs rise with each customer, transaction, unit of usage or service commitment?
4. Route to market
Can the chosen acquisition and sales motion be supported by the margin, contract value and retention profile?
Connect the model to the growth system.
A model is not only a revenue stream. It shapes pricing, acquisition, retention, sales motion, product design and the amount of operational effort required to deliver value. Use the research below to evaluate those links before optimizing isolated metrics.
How to evaluate a startup business model
Direct answer: evaluate a business model by tracing how customer value becomes revenue, how that revenue relates to delivery cost, what repeat behavior sustains the economics, and whether the chosen go-to-market motion can acquire and serve customers at a defensible cost. The model should be treated as a set of linked assumptions that can be tested and redesigned.
What is a business model?
David Teece describes a business model as the design or architecture of the mechanisms through which an enterprise creates, delivers and captures value. That framing is useful because it separates the model from a simple revenue label. Two companies can both charge subscriptions while having very different customer segments, cost structures, channels, retention dynamics and margins. Joan Magretta similarly distinguishes a business model from strategy: the model explains the economic logic of how the business works, while strategy also addresses how the company will perform differently from competitors.
The Business Model Canvas operationalizes this idea through nine connected blocks covering customer segments, value propositions, channels, customer relationships, revenue streams, key resources, key activities, key partnerships and cost structure. Strategyzer explicitly presents the canvas as a way to see how changes in one part of the system affect the others, rather than as a one-time planning document.
Why the revenue label is not enough
Labels such as SaaS, marketplace, subscription or freemium are useful shorthand, but they do not determine whether a company has attractive economics. A SaaS company can have weak retention and costly implementation. A marketplace can show transaction growth while struggling to retain supply. A subscription service can produce recurring billing without recurring customer value. The model becomes analytically useful only when the label is connected to customer behavior, revenue mechanics, cost-to-serve and the required sales motion.
This distinction also matters when founders borrow a model from a successful company. Similar surface mechanics can conceal different buyer power, regulation, infrastructure cost, switching friction or willingness to pay. TechStartupLabs therefore treats business-model analysis as a mechanism problem rather than a template-selection exercise.
Start with the customer-side mechanism
The first question is what job, outcome or recurring need creates willingness to pay. The answer affects whether recurring access, metered usage, transactions, licensing or another structure makes sense. It also affects packaging. A model that charges by user can work when each additional user receives distinct value, while a usage metric can be more intuitive when value rises with transactions, compute, storage or another measurable activity.
Customer concentration and buying process matter as well. A low-friction digital product sold to individuals can support self-service economics. A product that requires security review, procurement, implementation and executive approval may need larger contract value and stronger retention to support a high-touch enterprise motion. The model and the GTM system therefore need to be evaluated together.
Then test the cost and margin mechanism
Every revenue mechanism carries a delivery structure. Some software products have low marginal delivery cost after the platform is built, but implementation, support, infrastructure or AI inference can still scale with usage. Marketplaces may have relatively light production costs but significant spending on trust, payments, dispute resolution and supply acquisition. Licensing can scale efficiently in some cases, but negotiation, customization, compliance and enforcement can introduce substantial fixed or relationship costs.
The useful question is which costs scale with customers, transactions, usage or service obligations. If revenue grows but variable or support costs rise at the same pace, the apparent scalability of the model can be overstated. That is why model analysis should connect gross margin, acquisition cost, retention, expansion and payback rather than treating them as independent dashboard metrics.
Apply the framework to your company
Share the current customer segment, pricing structure, delivery model and growth constraint. The consultation can focus on the mechanism most likely to improve revenue quality or scalability.
Discuss a business-model decisionUse evidence to distinguish a model problem from an execution problem
A weak result does not automatically mean the business model is wrong. Low conversion can come from positioning or channel mismatch. Churn can come from poor onboarding, weak customer selection or a value gap. Slow enterprise growth can reflect procurement cycles rather than pricing architecture. Before changing the model, identify the observed signal, map the plausible mechanisms and determine which evidence would separate competing explanations.
This is also why a model redesign should be treated as a testable hypothesis. The Business Model Canvas itself is most useful when it is dated, revised and challenged as new evidence appears. A change in segment may require changes in channel, relationship, activities and cost structure. A change in pricing metric may alter buyer behavior, forecasting, margin and expansion. The decision is systemic.
Know when a hybrid model is justified
Many companies use more than one mechanism. A software company may combine subscriptions with usage charges. A marketplace can add subscription tools for suppliers. A licensing business can add implementation or support services. A freemium product can combine seat-based subscriptions with usage limits. Hybrids can improve alignment, but they can also make the proposition harder to understand and the revenue system harder to operate.
A hybrid structure is most defensible when each element solves a different economic problem. A base subscription can create predictable access revenue while a usage component captures expansion. A transaction fee can monetize exchange while a paid supplier tier funds premium workflow tools. Adding mechanisms only because competitors use them is weaker reasoning.
Business model, strategy and growth should remain connected
Academic work has long treated business models and strategy as related but distinct constructs. Casadesus-Masanell and Ricart describe a business model as reflecting the firm's realized strategy and distinguish it from the choices available under alternative contingencies. For founders, the practical implication is that the model should not be evaluated in isolation from competitive positioning, capabilities and the choices the company can sustain.
TechStartupLabs uses this connection to structure the wider site. Business-model pages explain the value and revenue mechanism. Pricing and monetization pages examine how payment is structured. Unit-economics pages test whether growth creates attractive contribution economics. GTM pages assess whether the sales motion fits buyer behavior and contract value. Growth and international pages then examine how those mechanisms change as the company scales or enters new markets.
Research sources
- David J. Teece, “Business Models, Business Strategy and Innovation,” Long Range Planning 43 (2010), 172–194.
- Joan Magretta, “Why Business Models Matter,” Harvard Business Review, May 2002.
- Strategyzer, Business Model Canvas, official tool and guidance.
- Ramon Casadesus-Masanell and Joan E. Ricart, “From Strategy to Business Models and onto Tactics,” Long Range Planning 43 (2010), 195–215.
Related TechStartupLabs research
The SaaS business model guide covers software-service economics, while the Subscription business model guide examines recurring access, renewal economics, billing cadence and retention across industries. The Freemium business model guide examines free-to-paid conversion and free-tier economics, while the Usage-Based business model guide covers consumption-based pricing and the Platform business model guide covers multi-sided network economics. The Marketplace business model guide covers liquidity, take rate, trust and two-sided transaction economics. The Licensing Business Model guide covers IP ownership, royalties, exclusivity, territory, field of use and commercialization trade-offs. The Advertising Business Model guide covers attention monetization, ad inventory, CPM/CPC/CPA structures, yield and advertiser demand. The Transaction-Fee Business Model guide covers event-based monetization, fee structure, ticket-size sensitivity and transaction contribution.
Related business and technology research ecosystem
Turn the model comparison into a concrete decision
Use the research to identify which mechanism is under pressure, then test alternatives against customer value, cost structure, pricing and GTM requirements.
Request a tailored model reviewUsage-Based Business Model
Explore how consumption metrics, variable costs and customer usage translate into revenue in the Usage-Based Business Model guide.