What does an AI meeting notetaker cost per user?
Turning a call transcript into notes, decisions and action items.
The short answer
On Claude Sonnet 5, at 15,000 input and 900 output tokens per call. Fifteen thousand tokens is roughly a forty-five minute transcript. Twenty calls a month is a person in about one meeting a day, which is conservative for the buyers of this product.
Your numbers
Now put your own in
The fields open with the profile above. Change them to match your feature and the read updates as you type.
Your numbers
Rough numbers are fine. You are checking whether you are near the line, not being exact.
The usual right answer for production features. Near-frontier quality at a fraction of the cost.
Everything you send it
What it sends back
How often one user uses it
What it costs you
What drives the cost
Transcript length, which is meeting length, which you do not control. A team that runs two-hour workshops costs three times a team that runs stand-ups, on the same seat price.
Where the estimate goes wrong
The seat price. This category has settled on low per-seat pricing, often under twenty dollars, and that leaves a small absolute budget. The feature frequently fails on the price rather than on the token cost, and no amount of prompt optimisation fixes a pricing problem.
What this does not price
The transcription itself is a separate bill, charged per minute of audio rather than per token, and it is not modelled here. On most meeting products it is comparable to or larger than the summarisation cost, so treat this figure as one half of the answer.
Other features people cost
Customer support agent
A conversational agent answering customer questions against your help centre and ticket history.
Retrieval-augmented search
Semantic search over your own documents, with a generated answer and citations.
Document summarisation
Turning long uploaded documents into a short summary or structured extract.
Next step
If meeting notes and transcription summaries is on your roadmap and the number here made you pause, that is exactly the conversation worth having before an engineer starts.