Choosing a Business Model
February 23, 2026
SaaS vs marketplace vs transactional vs professional services: how to choose the right business model for your startup based on stage, market, and constraints.
SaaS: Predictable Revenue, Long Ramp
SaaS — Software as a Service — charges customers a recurring subscription fee for ongoing access to software. The financial properties of SaaS are highly attractive: monthly recurring revenue is predictable, gross margins typically run 70–80% (software has near-zero variable cost per additional customer), and customer lifetime value grows with each additional month of retention. The defensibility comes from switching costs — once a customer integrates your product into their workflow, migrating to a competitor requires effort, retraining, and data export, which creates inertia that compounds over time.
The limitation of SaaS is the ramp. Most SaaS businesses show a J-curve: initial months have negative cash flow because customer acquisition costs are paid upfront while revenue accrues monthly. A customer who pays $500 per month and took $2,000 to acquire doesn't break even until month five. Scaling a SaaS business before unit economics are proven — before your payback period is below 12 months and your net revenue retention is above 100% — means growing a leaking bucket. Shopify and Notion illustrate the mature version of SaaS complexity: both use freemium entry points to reduce acquisition cost, and Shopify adds a transaction fee on payments to diversify revenue beyond the subscription.
Marketplace: GMV Commission and the Chicken-and-Egg Problem
Marketplace businesses take a percentage of each transaction between buyers and sellers on their platform. Airbnb charges approximately 3% from hosts and 14% from guests on gross merchandise value. Etsy charges 6.5% of the transaction price. The unit economics of a marketplace are compelling at scale: high gross margin (the platform adds no physical inventory), network effects that compound defensibility as more buyers attract more sellers, and cross-side network effects that create winner-take-most dynamics in well-defined markets.
The challenge is the cold-start problem: a marketplace with no sellers has no buyers, and a marketplace with no buyers attracts no sellers. Every successful marketplace solved this by manually seeding supply before launching demand. Airbnb's early team drove to San Francisco rental properties in person and helped hosts create listings. Uber's early team recruited drivers directly in each new city before opening the app to riders. The pattern is consistent: pick one side (usually supply), aggregate it manually through direct outreach, then activate demand. The commission percentage should be calibrated to the margin available in the underlying transaction — a marketplace charging 20% in an industry with 15% margins will not retain sellers for long.
Transactional and Professional Services
Transactional revenue models charge per transaction rather than per month. Stripe charges 2.9% plus $0.30 per successful card transaction; the model requires high transaction volume to generate significant revenue, but each transaction has near-zero marginal cost once the infrastructure is built. Transactional models lack subscription lock-in, meaning customers can switch to a competitor without any friction beyond integration effort, but they also have low barrier to entry — customers can start using a transactional product without a sales conversation or an annual commitment.
Professional services — selling time and expertise directly — are the fastest path to revenue for most early-stage founders and the hardest model to scale. Gross margins run 30–50% at best (people are the product and people have fixed availability), creating a revenue ceiling that grows only by adding headcount. The time-for-money ceiling means professional services businesses almost never reach $10 million in revenue with a small team. The strategic use of professional services is as a bridge: sell consulting or implementation work to generate cash and customer relationships, use those relationships to discover recurring software needs, and build the SaaS product that systematises the manual work you've been doing for clients. This "services first" path to SaaS is documented across hundreds of successful B2B companies.
Choosing Based on Stage and Constraints
With limited runway, professional services buy time — they generate revenue immediately without requiring product development. The risk is that services revenue is easy to confuse with product validation; clients paying for your time are not the same as customers paying for your software. Track these separately and resist the temptation to grow the services business indefinitely if the goal is a product company. Revenue-based financing from Clearco or Pipe provides an alternative for SaaS companies with strong retention: non-dilutive capital advanced against your monthly recurring revenue and repaid as a percentage of monthly revenue, typically at 5–8% total cost.
Network-effect businesses — marketplaces, social platforms, communication tools — require capital to solve the cold-start problem before they can become self-sustaining. This is the primary use case for venture capital: not to generate revenue faster, but to fund the period of operating at a loss while building the network density that generates defensibility. If your market has winner-takes-most dynamics and you have a capital-efficient path to network density, fundraising accelerates the outcome. If your market is fragmented with many sustainable competitors, bootstrapping to first revenue and raising a small seed to accelerate is a more capital-efficient approach than raising a large round on a speculative network-effects thesis.
Frequently Asked Questions
Can a startup have more than one business model at the same time? Yes, and many of the most successful ones do. Shopify combines a SaaS subscription with transaction fees on its payments product. Amazon combines e-commerce (transactional), Amazon Prime (subscription), and AWS (usage-based SaaS). The risk of multiple business models early is focus dilution — you're building two value propositions simultaneously before you've validated one. Start with the model that generates the most signal about product-market fit, and add complementary revenue streams once the primary model is proven.
Is freemium a business model? Freemium is a pricing strategy within a SaaS business model, not a business model itself. It uses a free tier to reduce customer acquisition cost and build a large user base, then converts a percentage of free users to paid plans. The conversion rate that makes freemium economically viable varies by product — Slack converts 28–32% of free workspace seats to paid, which is exceptionally high. Many freemium products convert 2–5% of free users to paid, which requires a very large free user base to generate meaningful revenue.
How do I decide between self-serve and sales-led growth? Self-serve works when the product can be understood and adopted without a human explanation, the contract value is too small to justify a sales call (below $10,000 ACV), and the buyer has authority to make the purchase decision without committee approval. Sales-led works when the product requires integration into complex enterprise systems, the buyer needs to convince procurement, legal, and security stakeholders, and the contract value is large enough ($50,000+ ACV) to justify a full sales cycle. Most B2B products start self-serve and add a sales motion as average contract value grows.
What is net revenue retention and why does it matter? Net revenue retention (NRR) measures how much revenue you retain from existing customers over 12 months, including expansion revenue and deducting churn and downgrades. An NRR above 100% means your existing customers are spending more over time — the business grows even if you acquire no new customers. Best-in-class SaaS companies (Snowflake, Datadog) show NRR of 130–140%. An NRR below 80% means the business is slowly losing its existing customer base faster than expansion, which is structurally unsustainable regardless of how fast new customers are acquired.
Should I start with annual or monthly pricing? Monthly pricing lowers the barrier to the first purchase; annual pricing improves cash flow and reduces churn. A common approach is to offer both with a 15–20% discount for annual. Early stage, monthly pricing generates more conversion data and faster feedback on churn patterns. Once you understand why customers churn and have addressed the most common reasons, push harder on annual pricing — it locks in revenue for 12 months and simplifies renewal forecasting.
