Usage-Based Billing: The Definitive Enterprise Guide

What Is Usage-Based Billing?

Usage-based billing charges customers based on what they actually consume. The bill goes up when usage goes up, and down when it goes down. It sounds simple enough, but the implementation behind it is anything but.

You’ve probably heard it called different things depending on the context. In telecom, it’s usually metered billing. Cloud providers like AWS popularized pay-as-you-go. Enterprise SaaS teams tend to use consumption-based pricing. Pay-per-use shows up in IoT and utilities. They all describe the same fundamental model: the customer pays for what they use.

The clearest way to understand how this differs from a subscription is with a real-world use case. For example, an AI company building on a large language model API gets billed $0.002 per 1,000 tokens. In January, when things might be quiet, they process 50 million tokens and pay $100. In March, after releasing a major new feature, usage goes up to 800 million tokens and the bill is $1,600. Their cost increased because their business grew. That’s the alignment subscription billing can’t offer.

Flat rate per unit

This is the simplest model. A fixed price is charged for every unit consumed, regardless of volume. Example: $0.0001 per API call. You make 500,000 calls, and you pay $50.

Tiered pricing

As usage goes over certain limits, the unit prices goes down. Example: the first million tokens are billed at $0.002 each, and anything above that drops to $0.0015. Customers who use more pay less per unit, which is a natural incentive to grow.

Volume pricing

Similar to tiered, but the rate that applies at the end of the period applies to all units, not just the ones above the threshold. Example: if a customer consumes 500GB, the $0.03/GB rate applies to all 500GB, not just the portion above the threshold.

Staircase / high-water pricing

The customer is charged at the rate of the highest tier they reach during the billing period. Example: if the pricing breaks at 50 units and a customer uses 51, all 51 are billed at the 51+ rate. This model is common in telecom.

Overage / burst pricing

A base subscription includes a set amount of usage, and anything over that amount is charged at an overage rate. Example: $500 per month includes 10 million tokens, and additional tokens are billed at $0.0002 each. It keeps revenue predictable while also allowing for growth.

Formula-based pricing

The price is calculated from a custom multi-variable formula. Example: (calls x unit rate) x (1 minus volume discount). This model can handle complicated business contracts where pricing depends on multiple factors simultaneously.

For a deeper look at how these models work, the comprehensive usage-based billing software guide covers the mechanics in detail.

Market Trends & Dynamics

Usage-based billing isn’t a new idea, but the rate of adoption has accelerated sharply. In 2018, 27% of SaaS companies used usage-based pricing. By 2023, that number was 61%, according to OpenView. OpenAI, AWS, Snowflake, Twilio, and Databricks all charge based on usage. It’s important to know why the most successful tech companies in the world all agree on a pricing model.

The biggest driver right now is AI. Every company building a product on a foundation model API faces a billing problem that subscription systems weren’t designed to solve. These problems include token-based pricing with different input and output rates, model-version-specific rate tables, and the need to handle a lot of usage events continuously. A traditional subscription platform bills once a month, but AI workloads generate billable events continuously. Because of that gap, a new wave of companies is moving toward consumption billing infrastructure.

The bigger change is from “pay for what you use” toward “pay for the value you get.” Hybrid models that combine a subscription base with usage-based expansion are now the most common enterprise pattern. Customers pay a set amount every billing period, and their bill goes up as their usage grows, without anyone having to renegotiate a contract. That’s a big change for CFOs. Usage-based revenue at scale is more predictable than it seems, especially when supported by minimum commitment structures and real-time usage analytics. Companies running this model well will usually see net revenue retention above 120%.

For a closer look at where SaaS pricing is heading, SaaS usage-based pricing trends and the consumption-based pricing guide cover the landscape in detail.

Benefits of Usage-Based Billing

Depending on which side of the invoice you’re on, the case for usage-based billing is different.

For the company selling the product:

The most immediate benefit is that there is lower friction acquiring customers. Deals go faster, and it’s easier for a new customer to say yes when they don’t have to make a big commitment up front. More importantly, the customer’s success leads to more sales, so there is no need for a separate upsell motion. If a customer’s usage doubles because their business is growing, the bill doubles automatically. Net revenue retention above 100% becomes a natural outcome rather than a goal to engineer. Customers who pay only for what they use are also less likely to cancel. The most common reason for subscription churn, “we’re paying for things we don’t use,” doesn’t apply.

For the customer buying the product:

The cost aligns with the value. A startup that processes 10,000 API calls pays a lot less than an enterprise processing 10 billion, which makes sense. There’s no large upfront commitment to get through procurement, which means faster time to value. And as the customer’s needs grow, pricing scales with them. There are no renegotiations or surprise renewal talks.

Finance leaders building the internal business case should read the CFO Playbook for Usage-Based Billing, which covers the financial model in detail.

Challenges of Usage-Based Billing

Usage-based billing is the right model for many businesses, but it can create operational complexity. Understanding the challenges upfront is what separates companies that implement it well from those that spend the following year fixing billing errors and dealing with customer complaints.

The most common mistake companies make with usage-based billing isn’t choosing the wrong pricing model. It’s treating billing as a finance problem when it’s actually a data infrastructure problem.

Revenue predictability is the challenge finance teams raise most often. Variable revenue is harder to forecast than a fixed subscription. The answer isn’t to avoid usage-based billing, it’s to build the right analytics, set minimum commitments where appropriate, and track usage trends closely enough to model scenarios with confidence.

At scale, billing is complicated. Different sources send usage data at different times and in different formats. Before they can become an invoice line, events must be ingested, checked for duplicates, changed, and rated according to pricing rules. When there are a lot of events, manual processes don’t work.

Bill shock happens when customers exceed usage thresholds without realizing it and receive an invoice they weren’t expecting. It’s one of the most common causes of churn in usage-based products, and it’s almost entirely preventable with real-time usage visibility and proactive threshold alerts.

Revenue leakage is less obvious but just as harmful. Dropped events, duplicate records, and transformation errors all reduce the likelihood of billing a customer. Companies that do usage billing without strong mediation often lose 1% to 3% of their revenue. At $10 million in annual recurring revenue, that could be up to $300,000 in sales that don’t show up on an invoice.

The ASC 606 and IFRS 15 effects come into play when it comes to recognizing revenue. Incorrectly recognizing variable usage revenue makes the company more likely to be audited. The billing system needs to send correct, time-stamped data to the revenue recognition engine. If billing and RevRec are in different systems, this handoff is always a risk.

None of these challenges are impossible to solve, but they all require the right platform infrastructure to address reliably.

Key Platform Requirements

Choosing a usage-based billing platform is an infrastructure choice that will affect your product, your finance team, your ERP, and your customers. It matters that you get it right. Below are the ten requirements that make a platform able to handle enterprise-grade consumption billing different from one that will cause more problems than it solves. Most of the time, failed usage-based billing implementations are due to either picking the wrong billing metric or not realizing how hard it is to get clean usage data from production systems into an invoice.

1. Real-time usage ingestion and mediation

Before it gets to the rating engine, raw usage data from production systems needs to be gathered, checked for errors, cleaned up, and turned into clean billable records. This is called mediation, and it stops billing mistakes from happening in the first place. The platform should ingest data from REST APIs, webhooks, CDRs, event streams, and file feeds, in near real-time. Without a mediation layer built into the billing platform, you’re adding a separate vendor and a new integration to maintain. BillingPlatform’s built-in mediation engine handles this natively.

2. Flexible rating engine supporting all pricing model types

Within the same contract, the rating engine must be able to handle all six pricing models: flat rate, tiered, volume, staircase, overage, and formula-based. Enterprise products often use more than one pricing model. For example, a SaaS platform might charge flat rate for standard API calls, tiered pricing for premium endpoints, and formula-based pricing for enterprise customers who have special agreements. Everything has to work at the same time without custom engineering. See BillingPlatform’s usage-based billing capabilities.

3. Hybrid model support

Pure usage-only pricing is the exception in enterprises. Most mature usage-based models combine a base subscription, which provides revenue predictability and a usage floor, with consumption-based overages that capture expansion. The platform needs to handle both in a single contract and consolidate them into a single invoice. Platforms that treat subscriptions and usage billing as separate modules force reconciliation work and create invoice fragmentation. See how BillingPlatform’s hybrid billing handles this natively.

4. Low-code or no-code pricing configuration

Prices change all the time because of sales tiers, volume discounts, new products, contract changes, and competitors’ actions. If every change to the pricing rules requires an engineering sprint, the finance and product teams won’t be able to keep up with the market as quickly as it moves. The platform should allow finance teams to configure new pricing models, modify existing rules, and launch new offerings through a visual interface without writing code.

5. Automated revenue recognition

Usage-based revenue is harder to recognize under ASC 606 and IFRS 15 than subscription revenue because the performance obligation is satisfied incrementally in variable amounts. The billing system needs to automatically send accurate, timestamped usage data to the revenue recognition engine. The best architecture is one where Rev Rec is built into the billing workflow instead of being handled by a separate module or third-party system. This is because every time systems have to talk to each other, there is a chance of reconciliation errors. BillingPlatform’s revenue recognition is built directly into the platform.

6. Real-time customer usage visibility

Customers should always be able to see their current-period consumption, projected invoice amount, and how their usage has changed over time. Without this, bill shock is inevitable. The platform should have a customer portal with real-time tracking, threshold alerts, and self-service capabilities. Finance teams need the same visibility with real-time dashboards showing revenue by product, by customer, and by pricing tier.

7. Multi-currency and global compliance

Enterprise usage-based billing is global. The platform needs to be able to handle billing in 50+ currencies with real-time FX rate application, handle VAT, GST, and sales tax in different jurisdictions, and support multi-entity account hierarchies for parent/subsidiary structures. Tax compliance built into the rating and invoice generation process is very different from tax compliance that is done after the fact. See BillingPlatform’s global support capabilities.

8. Dunning and collections automation

With usage-based billing, there are more invoices with varying amounts compared to subscription billing. This means that there are more failed payment events to deal with. The platform needs to automate retry schedules, escalating customer notifications, account suspension triggers, and collections handoffs, so that no one has to do anything manually at any point. BillingPlatform’s collections automation handles this end-to-end.

9. ERP and CRM integration

A usage-based billing platform sits at the center of the revenue stack. It needs to receive contract data from CRM systems like Salesforce, and post journal entries to ERP systems like SAP, Oracle, NetSuite, and Workday. Integrations should be native or API-based, not CSV exports that break on schema changes. BillingPlatform connects to your existing tech stack through native REST APIs, webhooks, and pre-built connectors.

10. Enterprise scalability and audit trail

There needs to be an unchangeable, time-stamped audit trail for every billing event, change in pricing rules, adjustment to an invoice, and revenue recognition entry. This is non-negotiable for companies subject to SOX controls or external audit. The platform should be SOC 2 Type II certified and capable of processing the transaction volumes your business will generate at 5 times your current size, not just where you are today. See BillingPlatform’s enterprise scalability.

For more on what to look for in a platform, the comprehensive UBB software guide covers evaluation criteria in depth.

Industry Use Cases

Usage-based billing isn’t just for SaaS. Most industries are adopting the model because the value a customer gets changes with how much they use it. Companies that use it well come from a wide range of fields, including software, infrastructure, communications, and physical services. You can see a range of examples on the BillingPlatform customer page.

SaaS and cloud software

Most of the time, SaaS companies need to charge for a mix of API calls, active seats, storage, and feature usage, all in the same contract. The billing problem is making a single, easy-to-read invoice that includes both a subscription and a consumption component, while also dealing with mid-cycle plan changes and making sure that variable revenue is compliant with ASC 606. A SaaS company charging $0.01 per API call plus a $200/month base subscription, for example, needs those two components rated, consolidated, and invoiced together, not reconciled manually across two systems at month end.

AI and LLM companies

AI products create a billing challenge that most platforms weren’t designed to handle: token-based pricing with different input and output rates, model-version-specific rate tables, and the need to support hybrid models where enterprise customers pay a subscription base plus token overages.

BillingPlatform’s AI monetization capabilities are purpose-built for this.

Telecom and Unified Communications (UCaaS)

Telecom and unified communications companies often charge by the minute, message, gigabyte, and DID number all at once when there are a lot of events. This makes the mediation challenge significant because call detail records arrive from carrier-grade systems in formats that need to be normalized before they can be rated. Billing based on usage, subscription pricing, and one-time charges often need to be on the same invoice.

Cloud infrastructure

Cloud infrastructure providers need to collect usage data across multiple regions, accurately meter burst workloads, and give customers visibility into what’s causing their bill. The billing model typically combines contract-based and usage-based components, with integrations into Salesforce and ERP systems required for clean quote-to-cash operations.

IoT and connected devices

IoT billing creates a high frequency of usage events, device interactions, sensor reads, data transactions, that need to be attributed to the correct customer and billing period at scale. The billing platform needs to handle this volume without performance degradation while supporting per-device or per-event pricing models.

Transportation and logistics

Transportation billing often involves formula-based rate plans that account for miles, weight, route segments, fuel surcharges, and contract terms simultaneously. Sydney Airport uses BillingPlatform to manage complex billing across airline movements, parking, retail, and more.

Usage-Based vs Subscription Billing

These two models are often described as opposites, but most enterprise companies end up using both.

The practical answer for most enterprise businesses is a hybrid model. A base subscription provides a revenue floor and gives customers a predictable monthly cost. Usage overages on top of that capture expansion as customers grow. The subscription component keeps the ARR predictable, which is what CFOs and investors need. The usage component means that revenue grows when customers are happy without having to make a new sale.

Most mature usage-based businesses have converged on this model, and it is much harder to change it later than it is to build it from the ground up. For a full breakdown of how to think through the choice, see the usage-based vs. subscription billing guide.

How to Implement Usage-Based Billing

Getting usage-based billing right is a cross-functional effort. Product, engineering, finance, and go-to-market all need to be involved.

The steps below are the right sequence to follow. Most of the time, the companies that get this wrong do it in the same way: they set the price and instrument the product before they fix the mediation problem. First, the data has to be clean.

Step 1: Define your billing metric

The billing metric is the unit of consumption that you’ll charge for. It needs to directly reflect the value the customer receives. If the metric goes up when the customer gets more value, you’ve got the right one. It also needs to be possible to measure it at the event level and link it to the right customer and billing period. Some examples are API calls, tokens, compute hours, active seats, and transactions processed. Don’t use metrics that are difficult for customers to predict or control.

Step 2: Instrument your product to capture usage

Every billable event needs to be tracked at the point of consumption. That means adding instrumentation to your product, call counters, event tracking, and feature flags tied to billing metrics. Every event record needs a customer ID, a time stamp, the type of metric, and the amount. The quality of the data here affects how accurate the billing is later on.

Step 3: Build or select a mediation layer

The raw usage data from production systems is almost never in the format your billing platform needs. A mediation layer collects events from multiple sources, validates completeness, removes duplicates, applies business rules, and changes the data into standardized billable records. For high-volume products, this layer needs to handle events in real time.

Step 4: Configure your pricing rules

The rating engine uses your pricing rules to figure out how much to charge once it has clean usage data. Set up your pricing model variants, tier boundaries, rates, overage handling, discounts, in the billing platform UI. Test every scenario before going live. When there are a lot of transactions, rating errors add up quickly.

Step 5: Connect revenue recognition and ERP

Usage-based revenue recognized under ASC 606 needs a clean, automated data flow from billing to Rev Rec and from there to your GL. Before the first live invoice, make this connection. Not after. Companies at this stage are most likely to have audit risks when they have to manually reconcile billing and RevRec.

Step 6: Build customer usage visibility

Deploy a customer portal with real-time usage tracking, threshold alerts, and projected invoice amounts before you go live. Customers who can see how much they use in real time don’t get shocked when they get their bill. That’s one of the best ways for a usage-based business to keep customers from churning.

Step 7: Test, launch, and iterate

Use real product data to run end-to-end billing tests. Test high-usage situations, edge cases, and currency changes. Check the output of the invoice. Check how often people use it every month after it launches. It’s normal for companies to change their pricing model at least once in the first year based on how customers actually act. This is not a failure.

For a detailed implementation guide, see best practices for usage-based billing implementation.

How to Choose a Usage-Based Billing Solution

Choosing a usage-based billing platform is a long-term infrastructure decision. The platform you select will be at the center of your revenue stack, touching your product, your CRM, your ERP, and your finance team’s close process. Think about where your business will be in 5 years, not where it is now.

The most expensive mistake in this evaluation is optimizing for the platform that handles your pricing model today, rather than the one that can handle the model you’ll need in two years when your product has evolved and your enterprise customers want custom terms.

1. Pricing model coverage

Does the platform natively support all six pricing models and hybrid contracts that include both a subscription and usage component? Can it handle formula-based pricing for complex custom agreements? Compare your current prices to the models you will probably need as your business grows.

2. Low-code configurability

Can your finance or product team configure a new pricing tier without having to open an engineering ticket? Ask for a live demonstration, not a recorded demo, of building a pricing model from scratch. If a developer needs to be in the room, that’s your answer.

3. Mediation and data ingestion

How does the platform handle event streaming, and is that frequency sufficient? Can it handle the data formats your product emits today? Ask for documentation on the mediation architecture, not just a checkbox on a feature list.

4. Revenue recognition

Is revenue recognition built into the billing workflow, or is it a separate module or third-party integration? How are ASC 606 revenue schedules calculated for variable consumption? Use a real contract example to walk through the RevRec workflow.

5. Scalability benchmarks

Ask for specific transaction volume numbers, peak load invoices per hour, and rating latency under sustained throughput. Ask for customers who are the same level as you for references. Marketing language about “enterprise-grade” performance is not a number.

6. Global and multi-entity support

How are multiple subsidiaries handled? What currencies are supported, and how is FX rate application managed? Which tax jurisdictions are covered natively?

7. Analyst recognition and customer references

Look at independent analyst coverage such as Gartner Critical Capabilities, Forrester Wave, and MGI 360 ratings. Then ask for references from companies in your industry at a comparable scale and complexity. A vendor with strong analyst recognition and referenceable enterprise customers has a much lower implementation risk than one without.

For detailed comparisons against specific alternatives, see how BillingPlatform compares as a Zuora alternative and as a Stripe Billing alternative.

How BillingPlatform Handles Usage-Based Billing

BillingPlatform is an enterprise usage-based billing platform that manages quote-to-cash for any consumption model, flat-rate, tiered, volume, overage, formula-based, and hybrid subscription-plus-usage, in a single cloud-native system. Recognized as a Leader in the 2025 Gartner Critical Capabilities for Recurring Billing Applications (ranked highest in 3 of 5 use cases), the Forrester Wave: SaaS Recurring Billing Solutions, Q1 2025, and #1 overall in the MGI 360 Ratings Report for Agile Billing, BillingPlatform serves global enterprises including Panera Bread, Clear Channel, Sydney Airport, DirecTV, net2phone, and Carrier.

Real-time mediation

BillingPlatform’s native mediation engine ingests usage data from any source, REST APIs, webhooks, CDRs, event streams, file feeds, validates and deduplicates events in real time, and transforms raw usage data into clean billable records. No separate mediation vendor required.

Configurable rating engine

All six pricing models are configurable through a point-and-click UI, with no engineering involvement required. Using an Excel-like expression syntax, finance teams can launch new pricing tiers, modify rate tables, apply discounts, and configure formula-based pricing. New pricing models can be launched in under 30 days. See BillingPlatform’s usage-based billing platform.

Hybrid model support

BillingPlatform natively combines base subscription fees, usage overages, minimum commitments, one-time charges, and milestone billing in a single contract and invoice. This is the key architectural difference from point solutions that are only used for usage billing. There’s no separate subscription module to reconcile against. See BillingPlatform’s hybrid billing capabilities.

Built-in revenue recognition

Revenue recognition is built into the billing workflow, not a separate module. Usage-based revenue is recognized automatically per ASC 606 and IFRS 15 rules as events are rated and invoiced. Finance teams can see real-time revenue schedules without manual journal entries or a separate reconciliation process. See BillingPlatform’s revenue recognition.

Multi-currency and global compliance

BillingPlatform can bill in more than 50 currencies with real-time FX, handles VAT, GST, and multi-jurisdiction tax natively. It also supports n-tier account hierarchies for parent/subsidiary enterprise structures. See BillingPlatform’s global support capabilities.

AI and token billing

AI products need a billing system that can manage token-based pricing, which includes distinguishing between input and output tokens, having specific rate tables for different model versions, and handling frequent usage events that happen. BillingPlatform supports these requirements natively, along with hybrid pricing structures where enterprise customers pay a subscription base plus token overages. See BillingPlatform’s AI monetization offering.

Enterprise scalability

BillingPlatform processes billions of transactions and billions of dollars every year for global enterprises. The platform is SOC 2 Type II certified and keeps unchangeable audit logs for every billing event, pricing rule change, and revenue recognition entry. See BillingPlatform’s enterprise scalability.

To see BillingPlatform’s full usage-based billing capabilities, visit the BillingPlatform usage-based billing solution page.

Competitive Landscape

There are several platforms that address usage-based billing at different points of the market, including Zuora, Stripe/Metronome, and Chargebee. BillingPlatform was designed specifically for enterprises that need to run multiple pricing models at the same time with revenue recognition built in. It doesn’t require separate modules or additional vendors for any part of the billing lifecycle.

For a detailed side-by-side comparison, see how BillingPlatform compares as a Zuora alternative and as a Stripe Billing alternative.

Analyst Recognition

BillingPlatform is ranked highest in 3 out of 5 use cases in the Gartner Critical Capabilities for Recurring Billing Applications, the most granular capability assessment Gartner publishes in this category, evaluating recurring billing applications across use cases including usage-based billing, hybrid billing, and multi-entity global billing.

BillingPlatform is named a Leader in the Forrester Wave: SaaS Recurring Billing Solutions, Q1 2025, recognized for its flexibility in handling complex pricing models and its low-code configurability across multiple enterprise verticals.

BillingPlatform received the highest overall rating (#1) in the MGI 360 Ratings Report for Agile Billing, evaluated across product functionality, customer success, go-to-market strength, and company viability.

If you’re evaluating usage-based billing platforms, request a demo to see BillingPlatform’s capabilities in your context. Prefer to explore on your own first? Start a guided tour of the platform.