Better Together: How Amazon Connect Blends AI With Human Empathy for Outstanding Support
Will AI replace support agents? No. The future is a blend: automate the simple, and let people handle what matters most-context, judgment, and empathy.
That "better together" model is how Amazon Connect frames modern customer service: AI handles the repetitive, humans handle the meaningful.
Why "Better Together" Works
Automate high-volume, low-value tasks so agents can focus on complex issues and emotional moments. As Pasquale DeMaio, VP and GM of Amazon Connect, put it, "No one is enjoying a password reset, neither talking to someone about it nor listening to the request." Let AI clear the queue; let people solve problems.
Result: faster resolutions on routine requests, and higher-quality conversations where loyalty is won or lost.
Empathy Is Human
"People aren't really looking for technology to form that emotional connection when they're trying to achieve an outcome," DeMaio said. A bot can say "I'm sorry," but customers can tell the difference.
His guidance to agents is simple: "Treat the customer on the phone like they're your friend." Not that they are your friend-treat them the way you would treat a friend. That's how empathy shows up naturally.
Support Doesn't Cost-It Pays
Chasing the lowest AHT can backfire. Use AI-driven self-service to clear the basics, then give agents the time and tools to build trust, reduce churn, and drive repeat business.
Proactive retention beats reactive escalation. With real-time analytics and sentiment detection, Amazon Connect helps flag at-risk customers so agents can step in early, personalize the fix, and save the relationship. Learn more about the platform here: Amazon Connect.
How to Balance AI and People in Your Operation
- Map top intents and automate the obvious. Start with the 10 most frequent requests (password reset, order status, appointment changes, billing copies).
- Set clear handoff rules. Escalate on repeated frustration, negative sentiment, authentication failures, or high-value accounts.
- Give agents context at a glance. Surface journey history, recent self-service steps, and key sentiment signals the moment the call or chat arrives.
- Use AI to assist, not replace. Offer real-time recommendations, knowledge suggestions, and compliance prompts-agents keep control.
- Coach empathy with simple cues. "Name the emotion, state the goal, outline the next step." Keep scripts short and human.
- Update scorecards. FCR, retention, customer effort, save rate, and LTV should sit next to AHT-not behind it.
- Close the loop with analytics. Mine transcripts to fix broken processes and deflect future contacts with smarter self-service.
- Make it easy to reach a person. Be transparent when a bot is speaking and provide a clear "talk to a human" option.
A Simple 90-Day Plan
- Days 1-30: Identify top intents, design self-service flows, define handoff triggers, and set your new metrics.
- Days 31-60: Launch self-service, enable real-time agent assist, and run empathy coaching with call reviews.
- Days 61-90: Tune prompts, refine routing, fix root causes from analytics, and report ROI (containment, CSAT, save rate, cost per resolution).
The Principle That Guides It All
"Think about the long-term value of the customer. And then think about how you would want to be treated as a human being. And then think about how AI can help you do that better." Keep that lens, and your tech choices get clearer.
If you're upskilling your team on AI-assisted support workflows, explore role-based learning paths here: Complete AI Training: Courses by Job.
Your membership also unlocks:
