← All use cases

Bell Canada · Montreal, QC

Bell Canada's Gemini-powered virtual assistants resolve over half of 16M+ chats on their own

Customer SupportProductivity

16M+ chat interactions, over half resolved by AI
5M+ voice assistant interactions
95%+ intent understanding
81% rated support effortless

What was done

Bell rebuilt its customer support chat and voice channels around AI virtual assistants running on Google's Gemini for customer experience, using customer data in BigQuery to give the agents context on each caller. A voice virtual assistant followed in the second half of 2025.

Outcome

Google Cloud's case study says the chat assistant has handled more than 16 million interactions with over half resolved without a human, the voice assistant has handled over 5 million, intent understanding is above 95%, and 81% of surveyed customers called the interaction effortless. Figures are vendor-published.

Evidence: Vendor case study (Google Cloud). Figures are as reported by the source and have not been independently audited by AgentsNorth.

Prompt Reconstructed by AgentsNorth

The source didn't publish its prompt. This is our reconstruction of a starting prompt for a similar workflow, not the company's actual prompt.

You are the virtual assistant for a national telecom provider's wireless, internet and TV customers. Use the customer's account context to understand why they're contacting us, confirm their intent in one short sentence, and resolve billing, plan, device and outage questions directly when our tools allow it. Never guess account details. If the customer asks for a person, is upset, or the request needs an action you can't perform, transfer to a live agent with a summary of the conversation so they don't have to repeat themselves.

Related use cases

Get new Canadian AI agent use cases each week

New use cases, one prompt to steal and one failure to avoid. Free. Unsubscribe anytime.