Customer Service · Engineering Blog · 🇨🇦 Canada · via agentusecases.ai
TELUS analyzes every call to fix problems before customers complain
The Canadian carrier anonymized conversations and used AI to find sentiment and root causes.
- Profession
- Customer Service Manager
- Tools
- Gemini Enterprise, BigQuery
- Source
- cloud.google.com ↗
- Country
- 🇨🇦 Canada
- Found via
- agentusecases.ai (Alex Finn) ↗
- Report date
- no date
- Verified on
- October 25, 2026
What they did
TELUS Digital moved chat, voice and video data into BigQuery, removed sensitive data and used Gemini to analyze the sentiment and root cause of each call. The findings feed agents that act before the customer complains.
How it went
This is a case study published by Google, the solution's vendor.
Good to know
Vendor case studies tend to highlight only the upside.
The prompt
Here are [number] anonymized support transcripts [paste]. Group them by reason for contact, identify the 3 most frequent root causes and suggest a preventive action for each.
Replace what is in [brackets]. Prompt written by us from the account, not a literal copy of the source. Do not paste other people’s personal data (privacy laws such as LGPD and GDPR apply).
Summary in our own words. Results as stated in the original account; we have not tested or audited them. Read the source ↗ Case found via agentusecases.ai (Alex Finn); text and prompt rewritten by us. Found a mistake?
Similar cases
01British car dealer group automates its customer service SMS
Stellantis &You UK closes out messages that need no action and flags unhappy customers.
02Clinic automates appeals of health insurance claim denials
An n8n workflow pulls the patient record when a claim is denied and AI drafts the appeal using the clinical data.
03Requesting an exchange for defective clothing from a photo
A photo of the torn pants, a receipt found in Gmail and an email to the store, ready to review.
04Agent navigates the insurer's phone menu until it finds a human
To change a policy, the user had the agent make the call and transfer only once a representative was available.
05Support triage with draft replies for review
A five-node workflow classifies each email, rates its urgency, summarizes it and writes a reply for a human to approve.
06Stopping your WhatsApp agent from duplicating customers in the CRM
A post shows why agents write to the wrong record: accents, invisible characters and stylized names.
07Adding an assistant that answers only from your WordPress site's content
A dev published a plugin with RAG and Gemini in the official directory that prevents made-up answers.
08Building an AI WhatsApp bot for small businesses
A developer built a customer service SaaS with Next.js and Baileys to avoid the official API's fees.