Pricing has changed quickly for software companies. In an a16z survey of 100 enterprise CIOs, L.E.K. Consulting summarized that 39% preferred pricing tied to tokens, API calls, or time. Hybrid models came second at 23%, and seat-based pricing trailed at 21%.
If you run a SaaS or AI product, the question is no longer whether customers want flexible pricing. It’s whether your SaaS billing software can support usage-based, seat-based, and hybrid pricing models as your business grows. Let’s walk through how to get there.
What Is Usage-Based Billing?
Usage-based billing charges customers based on measurable product usage instead of a fixed fee. The unit might be tokens processed, API calls made, minutes of compute or records stored. The more a customer uses, the more they pay.
How It Works
The process follows four connected stages: tracking usage, applying pricing, invoicing and reporting. Zenskar’s quickstart guide describes the same lifecycle in five steps. You add a customer, send usage events through an API, define billable metrics, set up a contract with usage-based products and generate the invoice.
Raw events are just records of activity. Billable metrics turn them into units you can price, such as total tokens per month.
Usage vs. Subscription vs. Hybrid
Each model fits a different situation.
- Subscription: A fixed recurring fee. It’s simple and predictable, and it suits products where value comes from team access rather than volume.
- Usage-based: Customers pay for what they consume. It suits products with variable costs, such as AI inference, where heavy users cost you more to serve.
- Hybrid: A base subscription plus metered usage. Software Equity Group data, cited by L.E.K. Consulting, shows hybrid companies averaging 19% revenue growth and 16% EBITDA margins, ahead of subscription-only peers.
Why SaaS and AI Use It
Value alignment is the first reason. Customers pay in proportion to the benefit they receive, which makes it easier to start small and grow. That flexibility removes a common barrier to buying.
AI costs are the second. Every model call consumes compute, so a flat fee can quietly turn a heavy user into a loss-making account. Charging by consumption keeps revenue tied to the cost of delivering the service.
Picture a startup that signs a customer on a flat plan, only to watch that customer run ten times the expected volume. It’s the kind of surprise that usually shows up with the cloud bill. Metering would flag it earlier.
Investors have noticed too. L.E.K. Consulting reports that consumption-based companies have traded at a valuation premium, though it expects that premium to narrow as the model becomes common.
Customers benefit too. They can scale usage up or down without renegotiating a contract. Does your current pricing give them that freedom?
Common Usage Metrics
The right metric reflects how customers experience value. These are the most common options:
- Tokens: Common for large language model products, where input and output volume drives cost.
- API calls: Suited to developer platforms and integrations.
- Compute time: Useful for processing, rendering or training workloads.
- Transactions: Fits payments, messaging and order-based products.
- Data volume: Works for storage, ingestion and analytics tools.
Choose a metric that customers can understand and verify. If they can’t predict it, they won’t trust the invoice.
Designing a Pricing Model
Once you’ve picked a metric, define how it converts into charges. Five building blocks do most of the work.
- Units: The measure you charge for, such as one million tokens.
- Tiers: Price bands that change the rate as volume grows.
- Included usage: A monthly allowance bundled into the base fee.
- Overages: The rate applied once a customer passes that allowance.
- Limits: Caps or alerts that prevent unexpected spikes.
For example, a plan could include a set number of API calls each month and charge a per-call rate beyond that. The numbers should come from your cost data and customer research, not guesswork.
Implementation Steps
A practical rollout usually follows this order:
- Select the metric. Tie it to customer value and your cost drivers.
- Set up tracking. Define your usage event schema first. Zenskar accepts usage through APIs, CSV uploads and data-source connectors, so teams with limited engineering time still have a way in.
- Define pricing rules. Turn raw events into billable metrics using no-code aggregation or SQL.
- Connect your systems. Link billing with your finance and data tools so invoices and ledgers agree.
- Test before launch. Run a few sample customers with known usage through a test setup, then check every invoice against a manual calculation.
Common Challenges
Usage-based billing software brings real operational demands. Real-time usage tracking, pricing complexity and ongoing price optimization are the hurdles teams most often plan around.
Bill’s predictability is a common concern. Spending alerts, usage dashboards and included allowances help customers see charges coming. Zenskar’s customer portal gives customers access to historical usage data, which supports transparency.
Data accuracy is the other big one. A missed or duplicated event can lead to an incorrect invoice, and incorrect invoices quickly erode trust. Reconcile usage against your source data regularly.
Forecasting also gets harder because revenue moves with usage. L.E.K. Consulting notes that hybrid models soften the volatility pure usage-based pricing can bring. The base fee provides a steady floor. L.E.K. Consulting also argues that execution matters more than the decision to adopt, so forecasting deserves attention early.
Billing and Revenue Recognition
Usage data, contracts, invoices and revenue schedules all depend on each other. The contract defines the pricing. Usage determines the quantity. The invoice records the charge, and the revenue schedule decides when that charge is recognized.
If these sit in separate tools, finance teams often end up reconciling spreadsheets at month-end. Zenskar translates usage and billing events into accrual-based journal entries for revenue, deferred revenue and receivables. Keeping everything connected cuts down that manual work.
Billing frequency matters too. Maxio found that 43% of SaaS billing software companies bill more often than monthly, which can improve cash flow.
Choosing Billing Technology
A billing platform should match how you price today and how you plan to price next year. Look at four things.
- Integrations: It should connect with your data warehouse, CRM, and payment and accounting tools.
- Flexibility: Look for subscription, usage and hybrid support, plus the ability to change metrics without rebuilding your setup. Zenskar’s decoupled architecture is designed for that.
- Scalability: It must handle growing event volumes. Zenskar describes near-real-time metering for AI businesses.
- Reporting: Finance and product teams both need clear usage and revenue views.
Conclusion
Usage-based billing works when three things are right: clear metrics, thoughtful pricing and accurate tracking. Start with one metric your customers understand, test it before launch and connect billing to your finance workflow. Are you ready to price the way your product is actually used?
