What does an AI phone assistant cost? Understanding the model

The cost of an AI phone assistant can only be reasonably assessed when it is clear what you are paying for. With webRichtung phone, you are not buying mere access to a tool. You pay in a Pay per Use model for work performed within a phone task that you define.

The released model

The account is free and has no base fee. When you self-register with an email and one-time code, you receive 7.5 Credits test budget. 1 Credit = 1 euro net. Once the test budget is used up, you load credit balance and pay for further usage from it.

This keeps access and work separate. The account itself creates no recurring base fee. Costs only arise when the platform performs a task area for you.

What Pay per Use practically means

Pay per Use is particularly suited to work volume that fluctuates. A business can test a narrowly defined phone task, evaluate the quality of results, and then decide whether this task area should receive more responsibility.

The relevant question therefore is not only: What does a call cost? More important is: What usable work results from it? An assumption is economically sensible only if the responsible person can then continue working with the information handed over.

Define a clear task before testing

Without a clear task, even a test budget says little. Define in advance:

• which call types the reception should handle
• what knowledge it may use for this
• what information your team needs for follow-up work
• which topics require a human to take over
• how you recognize a usable result

This way you test not the technology as a toy, but the assumed workforce in your own operation.

The overall calculation consists of work and follow-up work

An inexpensive call can be expensive if your team subsequently must request information, correct false expectations, or restart the entire process. Conversely, a cleanly bounded assumption can be valuable because it reduces interruptions and prepares the next step.

Therefore, alongside your Credit consumption, also evaluate the organizational effect: Do requests reach the right person? Is the needed information available? Does reception stay within its knowledge limits? Does unplanned interruption decrease?

No invented comparison calculations

Flat-rate market prices, per-minute models, or claims about the value of a won contract do not reliably help with your decision. Industries, call types, and follow-up work differ too much. A reliable calculation results from your actual task area and your own operational results.

Therefore, record what work reception assumes and what remains with the team. Only this distinction makes costs and benefits comparable.

Step by step rather than a grand cost promise

Use your test budget for a small, clear phone task. Test role, knowledge, boundaries, and handoff without unnecessary extensions. If the result holds up in daily operations, you load credit balance and expand responsibility incrementally.

This keeps financial control connected to operational management. You pay for completed work, not for access, and you decide for yourself which task area comes next. How to set up this role is explained in Setting up an AI phone assistant.