Use AI to handle customer calls before they escalate

TELUS used Gemini Enterprise to review every customer interaction, spot root causes, and guide staff with live recommendations.

Done with Gemini· Advanced

What they did
TELUS Digital brought chat, voice, and video call records into BigQuery, then removed private details with Sensitive Data Protection. Gemini Enterprise Agent Platform examined the full set for sentiment and recurring causes, and its findings were turned into Generative Playbooks for Gemini Enterprise for CX. The agents kept context across channels and supported frontline staff with real-time suggestions when a case moved to a person.
How it went
Google says 30% of contact center traffic is now handled through AI automation, and average resolution time dropped 87%. The case study also says the company reached $53.9M in yearly savings, and analysis expanded from roughly 1% of calls to all 22 million.
Tips
This approach depends on first combining channel data in BigQuery so the agent can see the full customer history. Treat vendor-reported performance figures as case-study claims and test the handoff rules carefully.

Try it yourself with Gemini

I want to build a Gemini-based agent that helps resolve [customer issue type] calls for [company]. Using the policy notes, transcripts, and examples I paste below, draft: 1) the agent’s goals and limits, 2) the signals it should look for, 3) the actions it can take without help, 4) the cases that must go to a human, and 5) a short set of real-world test cases. Before any action that would send a message, change a record, delete data, or submit anything, stop and ask me to approve it first. Notes: [paste here].

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on cloud.google.com ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.