Case studies / Logistics · Customer operations
How a cash-logistics group runs customer support with AI agents in 13 countries
One multi-agent system answers banks and retailers by voice, WhatsApp and email, configured for each country.

- 13countries on one agent core
- 150K+conversations a month
- 30–50%fewer human hours per use case
- 3channels: voice, WhatsApp and email
The situation
The group moves and manages cash for banks and retailers around the world. Its customers call, write and message about pickups, machine faults and orders, in many countries and languages.
The problem
Most requests need someone to look something up or change something in the operations and ticketing systems. Volumes are high, the requests repeat, and each country has its own language, rules and processes.
What we built
- One agent core
A single multi-agent system handles voice calls, WhatsApp and email, for customer service and for technical support.
- Configured per country
Language, rules and processes are set country by country, so adding a country is a configuration, not a new project.
- Connected to operations
Agents work inside the operations and ticketing systems, and pass the conversation to a person when the rules say so.
The results
- Live in 13 countries on the same core.
- More than 150,000 conversations a month.
- 30–50% fewer human hours in each use case the agents take on.
“Once validated, we took it to 7+ countries within weeks instead of years.”
Client anonymised. Full case file and reference available under NDA.


