Skip to content

Mission: Generate

Call Centers

In this episode, we explore how generative AI is a game changer for call centers and call center software companies (CCaaS). We explore real use cases and the technology to support them that demonstrates what’s possible with a state-of-the-art AWS-native solution.

Show notes:

AWS’s multi-part blog on integrating generative AI with call centers -

Our blog on how to deploy AWS native machine learning services, like Lex -

Official Transcript:


Welcome to Mission: Generate, the podcast where we explore the power of AWS and the cutting-edge techniques we use with real customers to build generative AI solutions for them. Join us each episode as we dig into the technical details, cut through the hype, and uncover the business possibilities.

I'm Ryan Ries, our generative AI practice lead and your host. In today's episode, we're going to talk about call centers. Okay okay, just wait! Don't turn me off just yet. I know it's not the most exciting sounding topic. But there is something exciting happening here, I promise!

If you've ever had to wait on the phone while lousy music plays, if you've ever been passed from one agent to another, if you've ever had to explain your issue multiple times--today's episode is for you. And if you're a leader in an E-commerce, Telecom, Travel, or Hospitality business, listen up, because you have a huge opportunity to up your customer experience.

Ready? Let's talk call centers.

A quick disclaimer before we kick things off. If you haven't tuned in yet, you might have missed our pitch for Mission: Generate.

But it's a podcast about Gen AI built with Gen AI. Which means that while what you hear may sound convincingly real, it is in fact a large language model operating on a script. Which makes me AI-Ryan, or RyAIn as we like to call him. Haha, awkward. RyAIn happens to be very useful because real Ryan is busy with customers and AWS. Far too busy to record a podcast.

Which is why every episode of Mission Generate was created this way. We hope it inspires you to see just what's possible with this technology. With that out of the way, let's talk call centers.

Let's start with the hard facts: call centers are a bummer. No one likes being on hold. No one likes having to call one in the first place. Nobody calls your customer support center because they're having a wonderful time. Something is broken, missing, or not working as expected. And that's why we call.

They're also a hard business problem, because the more you succeed the harder it becomes to run one. You'll get more calls, longer wait times, and more unhappy customers the more your business grows.

Amazon has made some great tools to meet this challenge. Services like Lex, Connect, Comprehend, and Contact Lens have already changed how call centers can scale. But now with AI, these tools can do a lot more for agents. We're entering a new era of customer support, where agents will be supplied a summarized context of previous conversations and given suggestions for resolution from AI-powered knowledge bases before they even pick up to answer. And our unsung heroes of phone support will be paired with AI assistants to help them get the right answer faster and more consistently, which means less time spent by us, the customers, re-explaining our problem or waiting on hold. A win-win!

This is also a major opportunity for businesses to go global, because you can leverage Generative AI to offer real-time translation of conversations. So if the agent is a non-native speaker, instead of struggling to understand your issue, they can have it translated into their language of choice as you speak.

And there's also an opportunity to improve the training of agents, because generative AI can simulate calls, too. Imagine a training course where you take calls from lifelike, unpredictable customers, even someone irate or hard-of-hearing, for example. This will better equip agents to handle unusual situations or problems they've never encountered.

Now, to run a Gen AI solution effectively for these problems, you're going to need speed and scale. Solutions need to be able to handle call volume at peak hours or when there's some type of incident, like an outage, affecting a lot of customers at the same time. But the major consideration is privacy. Sensitive call data needs to be carefully compartmentalized. Amazon's services, like Connect and Lex, give you those kinds of capabilities out of the box. But you'll also need to consider the security guarantees of the model you're passing data to. You also need to consider the adversarial caller--someone who thinks they can manipulate your automated system into giving them data that doesn't belong to them. The flip side is that there are also opportunities for improving security: AI agents can help avoid human intervention when critical, private information needs to be shared by a caller. So your identification procedure might be handled by an entirely automated system in the future.

Okay, right now, all of this might sound new or weird. Every breakthrough sounds that way at first. But imagine if a synthetic voice like mine picked up the phone and you could have a normal conversation with it. You could talk through your issue and get connected to the right person or even get it resolved instead of having to hunt through a phone directory, type your member ID in by hand, or yell for a representative until you're transferred.

Instead of dreading your conversation with an automated phone system, you'll be able to work with one that's sophisticated enough to resolve common problems or summarize everything your agent needs to know before you've even been connected.

Sounds like a better future, huh? That's the way the customer experience should be, and the businesses who get there first are going to see corresponding wins in their markets.

As you can see, it's a new era for call centers, and those that are powered by generative AI are the future of this kind of work. If any of what I just shared resonates with you or your goals as a business, you might have a great opportunity for improving your customer experience.

And if you're interested in doing that, but not sure where to start, reach out to us!

My team will give you an hour, free, just to chat about your use case. We're experienced at helping businesses scale and we take pride in building technical solutions for hard problems. So if you've got a problem that sounds anything like the ones we've just mentioned, or are just interested in Gen AI, drop us a line at mission cloud dot com.

I hope you enjoyed today's episode. And as always, good luck out there and happy building!

Subscribe to the Generate Podcast

Be the first to know when new episodes are available.