Single branch
Contact for self-serve pricing
- One branch
- The AI teller assistant
- Tasks, coaching, and required language during live conversations
- Retention controls
Community banks and credit unions
Napkin is an AI teller assistant.
It works alongside tellers and member-facing branch staff during live conversations.
- Tasks
- Coaching
- Compliance
The three things it does
01
Completes tasks on their behalf
Follow-up work is done before the member walks away.
02
Coaches them in real time on products and services
A newer teller knows which product fits the conversation in front of them.
03
Keeps their communication compliant as they speak
Required language happens in the moment, not in a post-hoc audit.
What it produces
- Work finished before the member leaves
- A newer teller who knows which product fits
- Required language in the moment
Notes may exist as a byproduct of the conversation. They are not the product.
Processed on Napkin’s own machines
Inference and processing run on infrastructure Napkin operates. There are no third-party cloud AI APIs. Capture is voice-only. Institution data is not used to train outside models.
- Processed on Napkin’s own machines
- No third-party cloud AI APIs
- Voice-only capture
- Data is not used to train outside models
A single branch, or the institution
Enterprise
Enterprise — custom
- Per-branch, billed with you
- The AI teller assistant across named branches
- Institution-set retention, roles, and audit access
- Pilot path available
Self-serve pricing is confirmed when you start a branch. Enterprise is quoted for the institution. Tax, hardware, and professional services are not included unless the plan says so.
Start with one branch
Self-serve is the default. A 90-day pilot and an enterprise conversation are available if you need them.