Usage metering
DEFINITION
Usage metering is the process of measuring how much of a product or service a customer consumes, in countable units like API calls or gigabytes stored, so that consumption can be billed accurately. It is the measurement layer underneath usage-based and hybrid pricing models.
TABLE OF CONTENTS
RELATED TERMS
Usage metering is the process of measuring how much of a product or service a customer consumes so that consumption can be billed. It captures countable events, such as API calls, gigabytes stored, messages sent, or seats active, and turns them into the quantities a usage-based or hybrid pricing model charges against.
Metering is the foundation under any pay-for-what-you-use pricing. Before a business can bill for usage, it has to record each unit reliably, attribute it to the right customer and billing period, and total it accurately at the close of the cycle. That involves collecting usage events from the product, aggregating them, handling corrections and late-arriving data, and passing clean totals into billing. A subscription platform such as Recurly can ingest usage records and apply them to a customer's plan and rates. Recurly supports two aggregation strategies: cumulative, which sums every usage record logged in a period, and last recorded, which bills only the most recent record. The strategy is selected per usage-based add-on, based on whether the metric represents accumulated activity or a current state.
Why usage metering matters for subscription businesses
As pricing moves beyond flat monthly fees toward usage-based and hybrid models, metering becomes the system of record that revenue depends on. If the meter is wrong, every invoice built on it is wrong, so accuracy here is not optional.
Metering matters for a subscription business because it:
Enables usage-based and hybrid pricing, which tie revenue to the value a customer actually gets.
Requires accuracy and auditability, since billed usage must be defensible to customers and finance.
Feeds not just billing but also product analytics and revenue reporting from the same event stream.
The harder parts are operational: deduplicating events, handling data that arrives late or out of order, and reconciling metered totals against what the product reports. How a platform handles those cases determines whether metering stays reliable or becomes a source of billing disputes.
How usage metering works
Metering runs as a pipeline from raw event to billable quantity. A typical flow:
The product emits a usage event each time a customer consumes a metered unit.
Events are collected and attributed to the correct customer and subscription.
Late-arriving data can trigger corrections to an unbilled period's running total; Recurly does not deduplicate usage records automatically.
Events are aggregated into totals for the current billing period.
At cycle close, the totals are rated against the plan's prices and tiers.
The rated amount appears on the invoice alongside any recurring charges.
Between cycles, near-real-time totals can drive usage dashboards and alerts so customers are not surprised by their bill.
Benefits and examples
Usage metering lets a subscription business price by consumption, bill accurately, and reuse the same usage data across product and finance. The main benefits are pricing flexibility, defensible invoices backed by recorded events, and one trustworthy usage dataset for both billing and analytics.
As an illustrative scenario, imagine an API product that meters calls and charges $0.002 per call, with the first 100,000 calls included in a $50 monthly base fee. A customer makes 250,000 calls in the month:
The first 100,000 calls are included and cost nothing extra.
Overage: 150,000 calls × $0.002 = $300.
Total: $50 base + $300 overage = $350.
These figures are hypothetical and only show how metered usage flows onto an invoice.
Frequently asked questions
What is usage metering? It is the measurement of how much of a product or service a customer consumes, recorded as countable events like API calls or gigabytes, so that consumption can be aggregated and billed.
What is the difference between usage metering and usage-based billing? Metering measures and totals consumption. Usage-based billing is the pricing model that charges for that consumption. Metering is the measurement layer the billing model relies on.
What kinds of usage can be metered? Anything countable and attributable to a customer, such as API requests, data stored or transferred, messages, transactions, compute time, or active seats. In Recurly this is defined through measured units, a required field on any usage-based add-on, configurable per plan or centrally on the Measured Units page, with no fixed list of supported unit types.
How are late or duplicate usage events handled? In Recurly, usage events are not deduplicated: two identical submissions are recorded as two separate records. Late or out-of-order usage, where the usage_timestamp falls in an already-closed billing period, is flagged as a correction. While the period is still unbilled, an accepted correction updates that period's running total. Corrections are rejected for add-ons using last-recorded aggregation or term-based billing, or if they would change the price tier of units already billed. Once a record is billed it is locked, and only its merchant_tag can still be edited.
