Imagine you’re tasked with managing the technology that supports overall owner satisfaction and customer retention for the world’s largest automaker. How can you enable AI to support your contact center operations, which are responsible for answering more than 1 million inquiries a year from drivers, prospective buyers, dealers, and field offices, without losing your customer loyalty base?
That’s the mission of Mitch Aubin, DevOps Product Manager for Consumer Technology Solutions at Toyota Automobile Dealers. This group works with customers, dealers and field offices to respond to inquiries and complaints. His group is working on long-term plans to integrate AI into its operations using stacks built on systems like Amazon Connect and Salesforce.
We spoke with Aubin and his colleague Bradley White, Toyota’s technology, innovation and automation manager, to discuss their progress so far. 4% of calls were handled through self-service, call handling time was reduced by 20%, and call transfers were reduced by 13%.
Editor’s note: This Q&A has been edited for clarity and brevity.
What problem did Toyota need to solve with AI and Amazon Connect?
Mitch Aubin
Mitch Aubin: Our guest. [customers] If you have any questions, please call Toyota’s Brand Engagement Center. It could be a bad experience at the dealership. It might just be a matter of how you connect Bluetooth to your car. “Hey, there was a recall last night. I learned about it on CNN.” Agents need to be knowledgeable about everything to provide answers as quickly as possible and ensure guests have a fun day.
We wanted to move towards a more omnichannel contact center and have the ability to route guests to the appropriate channel depending on the type of question they have. Different features were required, such as certain interactions being more sensitive and certain interactions best handled in chat with image sharing.
On-premises Genesys IVR has reached end of support [interactive voice response] We implemented a new Salesforce CRM to implement systems, manage cases, and track interactions. Brand Engagement Center was looking for a system that would allow agents to receive transcriptions of calls in real time so that they didn’t have to take extensive notes during calls. That was what they had to do. It was preventing them from being fully present during the call.
They also wanted a single screen so agents didn’t have to use a swivel chair for the many applications across Toyota that agents could use to resolve different types of customer issues.
How has the introduction of generative and agentic AI during transformation impacted your planning?
Aubin: We made a choice in 2023. We weren’t thinking about generative AI. Generative is a new technology that uses AI (natural language processing and machine learning) to give you the ability to learn about me and recognize my car model.
We have a traditional AI component that pays attention to who is calling and will likely recognize you as a guest. You have a Camry. Maybe they just issued a recall. We can notify them of the recall. Did you call because you wanted to know about it, or do you just want to know? You can also play it on your IVR and be connected to your preferred retailer to schedule a recall response appointment. So implementing traditional natural language processing AI was definitely on our roadmap.
Now, with generative AI technology, we also utilize real-time transcription of generative AI call summaries. With Amazon Connect Contact Lens, at the end of every call, you can capture a call summary directly into your Salesforce screen, recording details like “The customer called with this problem, the agent provided this answer, and the customer was satisfied,” or whatever the situation may be.
The agent can look at it and say, “Okay, that’s it.” You can add some bullet points if you want and save. The great value is that you get a standardized, jargon-free summary of the call. This means something completely different for someone seeing a call for the first time. Our legal team loves this because they can understand certain terms without needing a glossary to clarify them. Also, you don’t necessarily need to delve into real-time transcription, although it is possible.
What we don’t hear from you is that AI will replace humans.
Bradley White: Yes, that’s our approach. How can we achieve that human touch and enable them to do better?
By the time they contact us, the customer has already escalated the issue. It could just be an online search, not necessarily through Toyota’s channels. “I’m having trouble connecting Bluetooth.” or “I looked at your website, but I can’t figure out what new Land Cruiser or C-HR colors are coming next year. What colors will that car be available in?” or “My experience at the dealership didn’t meet my expectations.”
So they usually have contact information before contacting us. You don’t want to go back to full automation, right? We want to make agents more efficient by providing them with information so they can respond in a more personalized way.
Contents of the roadmap?
Aubin: Amazon is never static, so Toyota is a great partner for Amazon Connect. They are continually improving their roadmap and tools.
Initially, the goal was to reduce call forwarding volume by 5%, but a 13% reduction was achieved. This means customers are connected to the right expert within our Brand Engagement Center on their first call. Both our customers and our team are happier because we receive questions that our team knows the answers to and can take immediate action. Customers also love that they spend much less time addressing issues.
We will continue to enhance the IVR so that when people call, the IVR can answer questions and provide a great experience for guests. Toyota is looking at ways to use IVR to quickly resolve guest issues, rather than saying, “Hey, talk to me.”
Toyota is looking at ways to use IVR to quickly resolve guest issues, rather than saying, “Hey, talk to me.”
If they need to talk to a human, how do you get them on the phone with a human right away? Do you want a human to answer the phone right away? We want to provide a great guest experience so our customers can continue with their daily lives and our agents can assist customers who need their next attention.
I’ve heard that AI IVR works well for small businesses, but the larger the business, the more difficult it becomes. What does the world’s largest automaker need from AWS to improve its AI IVR?
Aubin: Everything is necessary to be able to handle multiple parties on a call at the same time. It’s possible. That will be taken care of today. But what we need help with, and what we’re working with them on, is to make sure transcription can support and identify with those multiple audiences. our guests, our dealers, ourselves, and [maybe] Another person on the call. You need to be able to clearly attribute what was said on the phone to each person who spoke so that you have a clear picture of what happened after the fact.
Don Fluckinger is a senior news writer at Informa TechTarget. He is responsible for customer experience, digital experience management, and end-user computing. Any tips? send him an email.
Mitch Aubin
