Introducing Usage-Based Billing for Enterprise

Our subscription plans cover a defined level of usage. For some applications, though, the resources needed can change considerably throughout a month. A product might create an inbox whenever a customer signs up. An agent might need an address for a task that lasts a few hours. A campaign might require hundreds of inboxes for two weeks, followed by a period of little activity.

Usage-based billing gives these organizations a billing model that accounts for those changes. Message charges reflect the number of messages processed, while charges for inboxes, domains, and storage account for how long those resources exist.

How billing works

There are five billable categories: messages sent, messages received, inboxes, domains, and storage.

Messages sent and received are billed individually. Inboxes and domains are billed for the time they exist. Storage charges depend on the amount stored and how long it is retained.

For example, if you create 500 inboxes for a two-week campaign and delete them afterward, their inbox charges cover those two weeks. If another set of inboxes remains active throughout the month, those inboxes are billed for the full period they exist.

An inbox does not need to be sending or receiving messages to accumulate inbox usage. Its existence contributes to the time-based charge, while messages are metered separately. Deleting the inbox ends its inbox charge.

This also means that your inbox count at the end of a month does not describe the entire month's usage. Resources created and removed earlier in the billing period still contribute for the time they existed.

Where this applies

Usage-based billing is intended for enterprise organizations that qualify based on volume. It is particularly relevant to applications that provision email resources as part of their normal operation.

Applications with an inbox for each customer

Some products assign a separate email address to every customer or end user. In these applications, inbox count changes as customers join or leave, and individual inboxes may handle very different amounts of email.

Usage-based billing accounts for the inboxes provisioned over time and the messages sent and received. Teams can also use inbox-level usage data to understand how consumption is distributed across their customers.

Agents that create inboxes for individual tasks

An agent may need an inbox for a negotiation, a verification process, or another job with a defined beginning and end. The address is useful while that work is in progress and can be deleted when the task is complete.

The same pattern applies to automated testing. A CI run might create an inbox to receive a verification email or magic link, then remove it during cleanup. Time-based inbox billing accounts for the period between creation and deletion.

Workloads with temporary increases in demand

Campaigns and batch jobs can require a larger number of inboxes for a limited period. An application may provision those resources at the beginning of a job, use them for several days or weeks, and remove them afterward.

In these cases, charges reflect the duration of the additional resources as well as the messages processed during that period.

Tracking usage

Usage data is available in the AgentMail console and through the Usage API. The API lets you retrieve usage over time, either for your organization or for an individual inbox.

For example, this request retrieves inbox and domain counts at hourly intervals:

curl -s "https://api.agentmail.to/v0/metrics/usage?start=2026-09-01T00:00:00Z&period=3600&usage_types=inbox_count,domain_count" \  -H "Authorization: Bearer $AGENTMAIL_API_KEY"

The response contains timestamped values for each requested metric:

{  "inbox_count": [    {      "timestamp": "2026-09-01T00:00:00Z",      "value": 1204    },    {      "timestamp": "2026-09-01T01:00:00Z",      "value": 1631    }  ],  "domain_count": [    {      "timestamp": "2026-09-01T00:00:00Z",      "value": 12    },    {      "timestamp": "2026-09-01T01:00:00Z",      "value": 12    }  ]}

Available metrics include inbox_count, domain_count, message_count, thread_count, and storage_bytes.

You can query a window within the last 90 days and adjust the interval with the period parameter. Use 60 for minute intervals, 3600 for hourly intervals, or 86400 for daily intervals. Shorter intervals provide more detail when examining changes during a job or campaign; daily intervals make longer-term patterns easier to review.

For time-based resources, both the count and its duration matter. Looking at the series over a billing period helps explain usage that a single snapshot would miss, particularly when your application frequently creates and deletes inboxes.

Usage by customer or task

If you provision inboxes on behalf of your own customers, you can retrieve usage for each inbox separately:

curl -s "https://api.agentmail.to/v0/inboxes/$INBOX_ID/metrics/usage?start=2026-09-01T00:00:00Z&period=86400" \  -H "Authorization: Bearer $AGENTMAIL_API_KEY"

By associating inbox IDs with customers in your application, you can use these records to attribute usage to individual accounts. This can support internal cost reporting or your own customer billing calculations.

The same approach works for task-based inboxes. If your application records which inbox belongs to each job, you can examine the usage associated with that job after it finishes.

Getting started

Usage-based billing is available to qualifying enterprise organizations and is enabled directly by our team. It is not currently self-serve.

To discuss eligibility and pricing, email support@agentmail.cc with a description of your application and expected usage.