# Revenue Models for SaaS: Subscription, Usage, and Hybrid Compared

> Compare every SaaS revenue model - subscription, per-seat, usage-based, credit-based, transaction-fee, outcome-based, freemium, and hybrid. Includes a decision flowchart and unit-economics breakdown.
- **Author**: Ayush Agarwal
- **Published**: 2026-04-21
- **Category**: Pricing, SaaS, Growth
- **URL**: https://dodopayments.com/blogs/revenue-models-saas

---

Your revenue model is not a pricing page decision. It is a product architecture decision that shapes how customers adopt, expand, and eventually depend on your software. Pick the wrong model and you either leave money on the table with power users or create so much billing friction that adoption stalls before you reach product-market fit.

Most SaaS founders default to monthly subscriptions because that is what they have seen work for other companies. But the SaaS landscape in 2026 is far more varied than it was five years ago. AI products carry real per-request compute costs. Infrastructure tools scale across orders of magnitude. Developer platforms need to charge for API calls, not headcount. The subscription-for-everything era is over, and the companies that recognize this early capture significantly more revenue per customer.

This guide breaks down every major SaaS revenue model, explains when each one works and when it fails, and gives you a framework for choosing the right one for your product. Whether you are building a vertical SaaS tool, an AI platform, or a developer API, the model you pick determines your unit economics, your expansion revenue, and your long-term defensibility.

## Revenue Models Compared

| Model | How It Works | Best For | Pricing Signal | Example Companies | Dodo Support |
| :--- | :--- | :--- | :--- | :--- | :--- |
| **Flat-rate subscription** | Fixed monthly or annual fee for access | Simple products with uniform usage | Time (monthly/annual) | Basecamp, Hey | [Subscriptions](https://docs.dodopayments.com/features/subscription) |
| **Per-seat** | Charge per user who accesses the product | Collaboration tools, CRMs | Headcount | Slack, Figma | Subscriptions with quantity |
| **Usage-based** | Charge per unit consumed (API calls, compute, storage) | Infrastructure, AI, developer tools | Consumption volume | Twilio, Snowflake | [Usage-based billing](https://docs.dodopayments.com/features/usage-based-billing/introduction) |
| **Credit-based** | Prepaid credits consumed by different actions | Multi-feature platforms, AI tools | Credit bundles | Midjourney, Jasper | [Credit-based billing](https://docs.dodopayments.com/features/credit-based-billing) |
| **Transaction fee** | Percentage or flat fee per transaction processed | Payments, marketplaces, fintech | Transaction volume | Stripe, Shopify Payments | Native (Dodo is MoR) |
| **Outcome-based** | Charge based on measurable results delivered | AI agents, performance tools | Value delivered | Performance marketing tools | Custom via webhooks |
| **Freemium** | Free tier with paid upgrades | Consumer apps, developer tools | Feature access or volume cap | Notion, Canva | Free plan + paid tiers |
| **Hybrid** | Platform fee + variable usage or per-seat | Enterprise SaaS, AI+SaaS combos | Base + consumption | Datadog, HubSpot | Subscription + usage meters |

## Flat-Rate Subscription

The simplest revenue model. Every customer pays the same monthly or annual fee for full access to the product. No usage tracking, no seat counting, no variable billing.

**When it works:**

- Products where all customers use the software in roughly the same way
- When the value proposition is access to a curated experience, not volume of use
- Early-stage products where simplicity in billing reduces friction during onboarding

**When it fails:**

- When power users consume 50x more resources than light users but pay the same amount
- When your cost structure has significant variable components (compute, bandwidth, storage)
- When enterprise buyers need customization that a single price point cannot accommodate

Flat-rate subscriptions remain effective for focused, opinionated products. But they are increasingly rare in B2B SaaS because they cannot capture expansion revenue as customers grow. Your revenue per account stays flat while your cost to serve often increases. For an in-depth comparison between recurring and one-time models, see [one-time vs subscription SaaS pricing](https://dodopayments.com/blogs/one-time-vs-subscription-saas-pricing).

## Per-Seat Pricing

Per-seat pricing charges based on the number of users who access the product. It scales revenue with organizational adoption and is the most common model in collaboration and productivity software.

**When it works:**

- Tools where every user interacts with the product directly (design, communication, project management)
- When buyer procurement teams have headcount-based budgets
- When usage intensity is roughly equal across users

**When it fails:**

- API products, infrastructure tools, or anything where headcount is irrelevant to value
- When buyers restrict seats to control costs, reducing adoption and increasing churn risk
- Products with a small number of power users and many occasional viewers

The per-seat model creates a predictable revenue stream but introduces a dangerous dynamic: customers optimize against you by minimizing seats. This is why many companies are shifting toward models that align price with value delivered rather than bodies in chairs. For a deep dive into seat-based mechanics, read [subscription pricing models](https://dodopayments.com/blogs/subscription-pricing-models).

## Usage-Based Pricing

[Usage-based pricing](https://dodopayments.com/blogs/usage-based-billing-saas) charges customers based on how much they consume. The consumption metric varies by product: API calls, compute hours, messages sent, data processed, or any other measurable unit that correlates with value.

> The billing system should disappear into the product. When you build on usage-based pricing from day one, your revenue scales with your customer's success. Retrofit it later and you spend six months migrating plans, re-educating customers, and patching edge cases in your invoicing logic.
>
> \- Ayush Agarwal, Co-founder & CPTO at Dodo Payments

**When it works:**

- Infrastructure and developer tools where consumption varies by orders of magnitude between customers
- AI products with real per-request compute costs
- Products where value directly correlates with volume (more API calls = more value extracted)

**When it fails:**

- When customers cannot predict their usage and need budget certainty
- When your consumption metric does not align with perceived value
- Products with high fixed costs and near-zero marginal costs per unit

Usage-based pricing is the fastest-growing model in SaaS, particularly for AI and infrastructure products. It naturally creates expansion revenue because customers pay more as they grow. The challenge is metering accuracy and invoice transparency. Customers tolerate variable pricing only when they trust the numbers. Learn more about implementation in our [usage-based billing implementation guide](https://dodopayments.com/blogs/implement-usage-based-billing).

For companies evaluating billing platforms that support metered pricing natively, see [best billing platform for usage-based pricing](https://dodopayments.com/blogs/best-billing-platform-usage-based-pricing).

## Credit-Based Pricing

Credit-based pricing gives customers a prepaid balance that gets consumed by different actions within the product. A basic API call might cost 1 credit while a complex AI generation costs 50 credits. The credit acts as an internal currency that abstracts away the complexity of variable costs.

**When it works:**

- Multi-feature platforms where different actions have wildly different costs
- AI tools where prompt complexity drives variable compute spend
- When you want to improve cash flow through prepaid purchases

**When it fails:**

- When the credit-to-value conversion feels opaque or confusing
- When credits expire before customers use them, creating resentment
- Simple products where a straightforward per-unit price is clearer

Credits solve a real problem: how do you charge fairly for a product that does ten different things at ten different cost levels? Instead of maintaining ten separate price lists, you define credit costs per action and let customers allocate their budget however they want. Dodo Payments supports [credit-based billing](https://docs.dodopayments.com/features/credit-based-billing) natively, with automatic deduction as customers consume your product.

For a broader look at how credits fit into AI product pricing specifically, see [AI pricing models](https://dodopayments.com/blogs/ai-pricing-models).

## Transaction-Fee Pricing

Transaction-fee pricing takes a percentage or flat fee from each transaction your platform facilitates. This is the standard model for payment processors, marketplaces, and fintech platforms.

**When it works:**

- Platforms that sit in the flow of money (payments, lending, insurance)
- Marketplaces where you facilitate transactions between buyers and sellers
- Products where your value is directly tied to the economic activity you enable

**When it fails:**

- When transaction volumes are low or unpredictable in early stages
- When customers process high-value, low-frequency transactions (a 3% fee on a $100K transaction feels very different from 3% on a $10 purchase)
- Products where the core value is not transactional

Transaction-fee models produce excellent alignment between vendor and customer: the more business the customer does, the more revenue you earn. But they require significant volume to generate meaningful revenue, which makes the early stages of a transaction-fee business particularly challenging. Operating as a [merchant of record](https://dodopayments.com/blogs/what-is-a-merchant-of-record) simplifies the compliance burden of handling money on behalf of others.

## Outcome-Based Pricing

[Outcome-based pricing](https://dodopayments.com/blogs/outcome-based-pricing-saas) charges customers based on measurable results: leads generated, contracts reviewed, revenue recovered, tickets resolved. It is the highest-alignment model because the customer only pays when they receive demonstrable value.

**When it works:**

- AI agents that perform autonomous work with clear, measurable outputs
- Products where ROI is quantifiable and attributable to the software
- When you have high confidence in your product's ability to deliver consistent results

**When it fails:**

- When outcomes are hard to measure or attribute (did the software cause the result, or would it have happened anyway?)
- When success depends on factors outside your control (customer data quality, market conditions)
- Early-stage products where performance is not yet consistent enough to guarantee results

Outcome-based pricing commands the highest margins in SaaS because the price is anchored to value, not cost. If your AI agent saves a customer $50,000 in labor costs and you charge $5,000, both sides feel like they won. The challenge is building the measurement infrastructure to track outcomes reliably. For a complete breakdown, see [outcome-based pricing for SaaS](https://dodopayments.com/blogs/outcome-based-pricing-saas).

## Freemium

Freemium gives customers access to a limited version of the product at no cost, with paid tiers unlocking more features, higher limits, or premium support.

**When it works:**

- Products with strong network effects where free users create value for paid users
- Developer tools where individual adoption leads to team and enterprise purchases
- When your marginal cost to serve a free user is genuinely low

**When it fails:**

- When free users consume expensive resources (compute, storage, bandwidth) without converting
- When the free tier is too generous and removes the incentive to upgrade
- When your product requires significant onboarding or support that free users still demand

The conversion rate from free to paid in most freemium products sits between 2% and 5%. That means 95% of your users generate zero revenue. Freemium works only when the 5% who convert pay enough to subsidize everyone else, and when free users contribute to growth through word-of-mouth, content creation, or network effects. For guidance on structuring your free tier, see [free trial vs freemium](https://dodopayments.com/blogs/saas-free-trial-vs-freemium).

## Hybrid Pricing

Hybrid pricing combines two or more models, typically a fixed platform fee plus variable usage-based charges. It is the dominant model for enterprise SaaS in 2026 because it solves the core tension between revenue predictability and fair value capture.

> The old debate was subscription versus usage-based. That debate is over. The answer for most scaling SaaS companies is both. A platform fee covers your fixed costs and gives the customer budget predictability. The usage component captures expansion revenue and keeps your margins healthy as adoption grows.
>
> \- Ayush Agarwal, Co-founder & CPTO at Dodo Payments

**When it works:**

- Enterprise SaaS with both platform value (dashboards, integrations, support) and variable consumption (API calls, compute, storage)
- Products transitioning from pure subscription to consumption pricing without disrupting existing customers
- AI products that want predictable baseline revenue plus compute-aligned upside

**When it fails:**

- When the platform fee component does not deliver standalone value (customers feel they are paying for nothing before usage kicks in)
- Simple products where a single pricing dimension is clearer and easier to sell
- When the split between fixed and variable is poorly calibrated, creating confusion

Hybrid pricing is where the SaaS industry is converging. It gives finance teams the predictable base cost they need for budgeting while ensuring that the vendor captures fair value as the customer's usage grows. For a detailed walkthrough of combining subscriptions with metered billing, see [subscriptions plus usage-based billing](https://dodopayments.com/blogs/subscriptions-usage-based-billing-saas).

## Which Revenue Model Fits Your Product?

Use this decision flowchart to narrow down the right model for your product type and customer profile.

```mermaid
flowchart TD
    A["What does your
product do?"] -->|"Facilitates transactions"| B["Transaction-Fee"]
    A -->|"Delivers measurable outcomes"| C{"Can you reliably
track results?"}
    C -->|"Yes"| D["Outcome-Based"]
    C -->|"Not yet"| E["Hybrid or Usage-Based"]
    A -->|"Provides tools or platform access"| F{"Does usage vary
significantly across customers?"}
    F -->|"Yes - orders of magnitude"| G{"Do actions have
different cost levels?"}
    G -->|"Yes"| H["Credit-Based"]
    G -->|"No - single metric"| I["Usage-Based"]
    F -->|"No - roughly uniform"| J{"Is adoption
user-driven?"}
    J -->|"Yes - each user interacts"| K["Per-Seat"]
    J -->|"No - team or org-level"| L{"Need predictable
+ growth upside?"}
    L -->|"Yes"| M["Hybrid"]
    L -->|"No"| N["Flat-Rate
Subscription"]
```

This flowchart is a starting point. Most companies refine their model over 2-3 pricing iterations as they learn how customers actually adopt and expand. The critical insight is that your revenue model should follow your cost structure and your customer's value perception, not the other way around.

## Unit Economics by Model

Each revenue model creates a different unit-economics profile. Understanding these patterns helps you forecast growth and identify when a model is working or breaking down.

**Subscription (flat-rate or per-seat):**

- CAC payback: 6-18 months typical
- Gross margin: 80-90% (low variable costs)
- Expansion revenue: limited to seat growth or tier upgrades
- Churn impact: high, because lost accounts represent 100% of that revenue

**Usage-based:**

- CAC payback: Variable (depends on ramp speed)
- Gross margin: 60-75% (variable compute/infra costs)
- Expansion revenue: strong natural expansion as customers grow
- Churn impact: gradual - customers often reduce usage before fully churning, giving you a warning signal

**Credit-based:**

- CAC payback: Fast (prepaid credits improve cash position)
- Gross margin: 65-80% (depends on credit-to-cost mapping)
- Expansion revenue: strong if customers consistently repurchase
- Churn impact: deferred - customers may stop buying credits months before you notice

**Hybrid:**

- CAC payback: Moderate (platform fee provides baseline)
- Gross margin: 70-85% (blended fixed + variable)
- Expansion revenue: strong from both tier upgrades and usage growth
- Churn impact: moderate - you retain platform fee revenue even if usage dips

For a deeper look at enterprise pricing economics specifically, read [enterprise SaaS pricing models](https://dodopayments.com/blogs/enterprise-saas-pricing-models).

## How to Migrate Between Revenue Models

Changing your revenue model mid-flight is one of the hardest operational challenges in SaaS. But it is sometimes necessary when your original model no longer fits your product or customer base.

**From subscription to usage-based:**

1. Start by adding a usage component alongside the existing subscription (hybrid first)
2. Grandfather existing customers on their current plan with a sunset date
3. Make the usage-based option clearly better for new customers (lower entry price, pay-for-what-you-use messaging)
4. Migrate existing customers in cohorts, starting with those whose usage already exceeds their flat-rate value

**From freemium to paid:**

1. Identify which features free users depend on most
2. Move those features behind a paywall gradually (not all at once)
3. Offer a generous discount for early converters
4. Keep a meaningful free tier - killing it entirely destroys your acquisition funnel

**From per-seat to hybrid:**

1. Introduce a usage metric that captures value beyond headcount
2. Reduce the per-seat price while adding a usage component, keeping total cost roughly flat for average customers
3. Communicate the change as "fairer pricing" - heavy users pay more, light users pay less

For billing infrastructure that supports multiple models and model transitions without custom engineering, see [best subscription billing software](https://dodopayments.com/blogs/best-subscription-billing-software). Dodo Payments supports subscriptions, usage-based billing, and credit-based billing in a single platform, which means you can run hybrid models or migrate between models without switching vendors.

## Building Revenue Infrastructure with Dodo Payments

Regardless of which revenue model you choose, the billing infrastructure underneath it needs to handle:

- **Metering**: Accurate, real-time tracking of usage events for consumption-based models
- **Subscription lifecycle**: Upgrades, downgrades, cancellations, proration, and renewals
- **Global tax compliance**: Automatic VAT, GST, and sales tax calculation across 220+ countries
- **Payment processing**: Credit cards, local payment methods, and multi-currency support
- **Merchant of Record**: Dodo acts as the legal seller, handling compliance and remittance so you do not have to

Dodo Payments is built for SaaS companies that need to support complex revenue models without building billing infrastructure from scratch. Whether you run pure subscriptions, usage-based pricing, credit systems, or any hybrid combination, the platform handles the billing mechanics while you focus on the product. See [recurring payments guide](https://dodopayments.com/blogs/recurring-payments-guide) for subscription setup, or explore [metered pricing](https://dodopayments.com/blogs/metered-pricing-guide) for usage-based implementation.

Get started at [dodopayments.com](https://dodopayments.com) or review [pricing](https://dodopayments.com/pricing).

## FAQ

### What is the most common revenue model for SaaS companies in 2026?

Hybrid pricing (a fixed platform fee plus variable usage-based charges) is the fastest-growing model for B2B SaaS. Pure per-seat subscriptions are declining as companies look for ways to capture expansion revenue that scales with customer adoption rather than headcount.

### How do I choose between usage-based and credit-based pricing?

Use usage-based pricing when your product has a single, clear consumption metric (API calls, compute hours, messages). Use credit-based pricing when your product performs multiple actions with different cost profiles and you need a single currency to abstract that complexity. Credits also improve cash flow because customers prepay.

### Can I combine multiple revenue models in one product?

Yes, and most scaling SaaS companies do. A common pattern is a base subscription for platform access plus usage-based charges for consumption above an included threshold. Dodo Payments supports subscriptions, usage-based billing, and credit-based billing natively, so you can run hybrid models without custom billing engineering.

### When should a SaaS company switch its revenue model?

Switch when your current model creates persistent misalignment. Warning signs include: customers gaming seat counts to avoid paying, power users consuming far more than they pay for, churn spikes at renewal because customers feel overcharged for light usage, or stagnant expansion revenue despite growing adoption within accounts.

### What revenue model works best for AI SaaS products?

Most AI products use either credit-based or usage-based pricing because AI inference carries real per-request compute costs. A flat subscription creates margin risk when heavy users consume expensive GPU resources. Credits or usage-based models align your revenue with your cost of goods sold. See our detailed breakdown in [AI pricing models](https://dodopayments.com/blogs/ai-pricing-models).

## Final Take

Your revenue model is a structural decision, not a marketing one. It determines how your company captures value, how customers perceive fairness, and how your unit economics evolve as you scale. The best model is not the one that sounds most sophisticated. It is the one where your revenue grows naturally as your customers succeed.

Start with the model that matches your current cost structure and customer profile. Plan for the model you will need in two years. And build on billing infrastructure that lets you evolve without starting over.
---
- [More Pricing articles](https://dodopayments.com/blogs/category/pricing)
- [All articles](https://dodopayments.com/blogs)