This article is in conjunction with the following two articles:
Read these two articles first to get the most value from this article.
Issues with native billing provided by voice providers.
#1 Voice providers (like Retell AI) show usage and billing at the workspace level, not per client or per agent.
So if you're serving multiple clients from one workspace, you're stuck:
- Exporting logs.
- Calculating minutes and token scaling per client/agent.
- Reconciling everything in spreadsheets.
- Creating invoices manually.
- Chasing payments.
This doesn't scale and is prone to errors.
#2 You can use a separate workspace for each client to separate usage and billing per client. But you still can't apply a markup on top of your Retell raw costs, which is required because Retell billing does not reflect the true cost of running a voice agent.
Costs like custom telephony, maintenance fees, and API call costs to external systems aren't reflected in Retell usage and billing.
So even with multiple workspaces, you would still need to create invoices manually.
#3 Voice providers like Retell do not allow you to create branded invoices, help protect your margins, or let you hide the underlying infrastructure (your prompts, error handling logic, test cases, telephony, etc.).
Once you give a client direct access to their workspace, they are inside your Retell account and can see everything from raw cost to the underlying infrastructure. There is no native way to present this under your own brand.
Here's the thing: Professional voice AI agencies don't do any of this.
They don't export logs, manually calculate minutes and token scaling per client/agent. They don't create and send invoices manually.
Want to know how? Check out my full breakdown in the article.