Calls to customer support lines are usually a test of patience and often don’t end well, but Mikhail Naumov (@MikhailNaumov), author of AI Is My Friend, is here to change that using artificial intelligence.
In this episode of Author Hour, Mikhail shares what he’s learned as the co-founder and president of Digital Genius, the leading platform for making customer service calls a little more human using AI.
Listen in to Mikhail to learn:
- The future of customer support
- Lessons learned while working with dozens of contact centers
- How companies are using AI to scale faster, make employees better at their jobs, and keep customers happy
How did you first recognize that AI has the potential to change the entire customer support experience?
I’ll never forget when we launched our very first chatbot for BMW. They were an early customer of ours, this is going back three and a half years; at the time we were still a couple of guys sitting in a not-so-nice office with only one mission: To build technology that helps companies automate conversations with their customers.
Early on we thought that the best way to do that was through automated chatbots that would be able to have meaningful conversations with customers on behalf of the brand.
At the time, BMW was launching their electric cars in the UK, the I3 and I8, so part of the message they wanted to get across was that their marketing for these vehicles was going to be as innovative as the car itself.
So, they hired us to come up with a chatbot that could carry a coherent conversation with a potential customer. That was our first foray into helping large brands, successful enterprises, have technology-infused conversations with their customers.
The campaign started with TV ads which would showcase these new cars and at the end of every single spot, there was a phone number that anyone could text with a question about the car.
For example, you could ask, “How much does it cost? Where can I buy one? or How long does it go on one electric charge?” Potential customers were able to ask fairly basic fact-based questions but they were also able to have more complex conversational experiences like scheduling a test drive or signing up for dealership visit.
That campaign lasted more than a year and hundreds of thousands of people texted that number after seeing those TV spots. So, it was a great tool for BMW to be able to engage with their customers without necessarily having to hire thousands of people to have these conversations in real time.
For us, it was an amazing learning experience. We learned that chatbots, while they’re useful in very narrow-use cases, such as answering basic questions about a car, telling you the weather, or helping you order pizza, they’re not great for customer service.
“It was after working with BMW that we realized we needed to move away from chatbots and that the customer service industry was ripe for disruption.”
So we set out to build a technology that could actually help individual customers with unique support issues on a case-by-case basis.
How do you move from chatbot to fully autonomous customer support?
Chatbots get a lot of attention, they’re quite hyped up, but it’s important to remember that a chatbot is just a title. What you really need to do is look under the hood and figure out what’s powering these bots.
In most cases today, chatbots are powered by a traditional form of natural language processing (NLP) which essentially relies on someone sitting down and writing out a bunch of scripts manually. The bot then uses keywords in the questions that people ask to choose a suitable response.
Somebody would have to sit down and think of every way possible that customers could ask a question about a product or service. Then they would try to match those questions to the appropriate answers. As you could imagine, this process is very lengthy and arduous, it takes a lot of time, and it doesn’t really scale well, especially in an environment like customer service where each unique question is so different from the other.
That’s one of the reasons why chatbots based on NLP don’t really work in customer service, and it’s a lesson we learned from working with BWM and some of our other early clients.
To move toward autonomous customer service we took the entire backend of our technology and reworked it from basic NLP to full-force machine learning and deep learning technology.
What are machine learning and deep learning?
Deep learning is a subset of machine learning where essentially you use technology to digest massive amounts of historical data, these data are then used to process inputs and create outputs. In the world of customer support, the inputs are questions that customers are asking and the outputs are answers that exist inside of a contact center.
Every time you have a problem with your bank or with your airline, for instance, you write in an email or make a phone call; we consider that an input. On the other end of that input is an answer that you get from the company or from the company’s representative; that’s an output.
Believe it or not, in those contact centers there are millions upon millions of historical customer service conversations stored in the format of transcripts and data logs, so we can take that data and feed it into a deep learning algorithm which converts all of these inputs and outputs into numbers and equations.
Those equations can be used to create a model which then predicts an output from a given input.
“The important difference between deep learning and an NLP chatbot is that deep learning is automated; we don’t have to actually go in and manually script all the different ways someone can ask a specific question and hope that the bot gets it right.”
With deep learning, we can rely on real, historical conversations that have already taken place to train the computer model. The fact that deep learning is based on authentic human interactions makes it a lot more resilient, a lot easier, and much more cost effective to implement than a traditional chatbot.
Can you tell us a bit more about your book, AI Is My Friend?
Yeah, I’ve spent the last three and a half years of my life living inside contact centers. I’ve literally spent thousands of hours sitting down next to hundreds of agents and watching what they do every day. I’ve watched them do their jobs, I’ve talked to them about the various things they have to do, and I’ve learned about the types of tools that they have to use.
I’ve done all of this to figure out one thing:
“How do we make the latest advancements in machine learning practical and useful in the contact center environment?”
After three and a half years we learnt a lot, so we decided to put all of these insights into a book: AI Is My Friend: A Practical Guide For Contact Centers.
It’s really written as an underground guide or as a guide from the trenches for customer service experts, practitioners, and leaders who have heard about AI and understand its importance but aren’t quite sure how to actually implement it.
This book is there to help you do just that. It addresses the entire ecosystem of AI, it’ll help you figure out how to pick the right partners to work with, it dispels any of the myths surrounding AI, and it’ll walk you through how to actually put this stuff into place.
What are the key points that someone has to know about AI before they decide to implement this in a call center context?
“The first thing to understand is that you have to differentiate between what AI is and what AI is not.”
First and foremost, AI is not there to replace your contact center overnight. You won’t be able to shut down your call center tomorrow once you implement AI, that’s just not how it works.
The second thing to understand is that there are a lot of companies out there claiming to be AI experts, but once you look under the hood, all they do is a bunch of traditional rule-based scripting that has been written to try to mimic what artificial intelligence can do. It’s not the same thing.
What is AI and what isn’t AI? Asking that question up front is really important and it’s something that the book gets to the root of very quickly so that when you see something in front of you, you’ll be able to tell if its AI or if it’s something trying to mimic AI.
Part two is understanding the spectrum of AI and that spectrum ranges from low-level intelligence applications, things like scripted chatbots, to the high-level stuff you see in Hollywood movies. I call that Hollywood AI. A lot of the AI you see in movies is obviously fiction, so although there may be a lot of hype around them, they’re really not becoming a reality anytime soon.
In the middle of the spectrum is an area we like to call practical AI. This is the technology that you can actually implement today to make your business run more efficiently.
Let’s say I’m a skeptical contact center manager, why should I listen to Mikhail Naumov? Why is AI important?
There are a number of reasons why it’s important to implement AI and figure out an AI strategy for your contact center right now.
The first thing is that inside of a customer service environment, we’re always facing the same challenges. The volume of queries is growing, customers have higher expectations for better service than ever before, and contact center budgets certainly aren’t getting any bigger.
The reality is, a lot of contact centers today are being portrayed and perceived as a cost center to a business. You’re constantly under pressure to try to deliver more for your customers with a smaller budget or with fewer resources.
Well, what happens when a new channel comes out like Facebook Messenger and suddenly, your customers are expecting to be able to talk to your company and your brand through Facebook Messenger?
This exact scenario just played out with one of our clients, KLM, Royal Dutch Airlines. They became the first airline to serve their customers via Facebook Messenger and WhatsApp. Unexpectedly, the volume of messages that KLM received through these new channels was astronomical and that’s very difficult to plan for.
The only way to really solve that problem in the past was to hire more people. Now, we can come in and implement an AI-augmented solution to reinforce your existing team of agents.
We can give them the right tools so they can perform at their best while the AI technology is automatically trained using all the historical conversations that have already taken place in your contact center. As new queries and questions come in, the AI is able to help your agents answer those questions faster.
“Improving your agent efficiency is one benefit, but overtime is AI also learns from the agents themselves to the point that it can begin automating some repetitive questions and answers.”
For companies like KLM and many others that have implemented Digital Genius software, or just AI software in general, what they are looking for is to scale their contact center in a cost-effective way and invest in their people by providing them with the best tools. I don’t know why a contact center manager wouldn’t want to do that.
Can you share some of the results your clients have seen from implementing your AI solutions?
First of all, AI isn’t there to replace your contact center. AI is there to reinforce your existing contact center so you can scale it more efficiently and more quickly. How does it do this? First, AI helps your agents be more productive on a day-to-day basis.
Maybe it used to take your agents 3, 5, or 10 minutes to handle an incoming email, or a call, or a live chat conversation, well AI drives that time down so that they can handle these questions a lot faster.
“AI ensures customer service agents don’t have to even think about the repetitive steps associated with taking a call or responding to an email that are normally performed manually.”
What we’ve been able to do for clients is help them boost their customer service metrics. Things like average handling time, or how long it takes for an agent to handle your case; CSAT, or customer satisfaction, how satisfied customers are with the overall experience that they’re getting; or first response time, how long do you have to wait until you get a satisfactory answer to your question.
These are all important metrics for contact centers and our AI solutions directly address these issues.
Now, when it comes to talking about some specifics, we have some great customers. Our software is already powering over 30 contact centers globally. Among them, we have KLM, Royal Dutch Airlines, which I mentioned earlier.
In KLM’s case, we have significantly reduced their average handling time by about 35% which allows their agents to handle more cases and spend a little more time on those really sensitive cases where their agents can now go above and beyond in terms of providing an exceptional customer experience.
Five years from now, where do you think Digital Genius will be?
The goal for the next few years is to continue establishing relationships with our current and prospective clients. We’re focused on being the market leader in AI for customer service.
You’ve got the situation now where AI is so hyped up and everybody is talking about it but it’s not quite clear yet who’s just talking the talk and who’s actually delivering results.
“We’re fortunate in that the companies we work with now are seeing results and they’re sharing those publicly through case studies and through press releases. That feels good but this is really just the beginning.”
Five years from now, I want to make sure we maintain our market leadership in AI for customer service but I also believe that by then, we’ll be branching even beyond the customer service space.
So, while today we remain focused on contact centers, the trend in the customer service world is that contact centers are very quickly evolving from being a place where people only seek answers when something goes wrong to the first point of contact between a customer, or even just a potential customer, and a brand.
Agents will need to be able to do far more than just answer basic questions and solve technical problems. They’ll be representing the brand at all stages of the sales funnel. When that happens we’ll be ready to provide our clients with the tools they need to help their agents do that job.
What’s one thing someone listening can do this week to prepare for the increasing influence of AI in business?
One of the most important components of your job, or even in your career at large, is to be educated at every stage of the game. The world is evolving today faster than it ever has before.
Think back to our parents and our grandparent’s day and age, the rate of technological change was much slower. New technology wasn’t coming out every year, or every month, certainly not every day.
Now, you’ve got new tools coming out by the minute, by the hour, so in a cutting-edge industry like artificial intelligence, it’s really important to be educated.
“My challenge to listeners today is simple: Go out there and read up on AI, read articles about deep learning and neural networks.”
Try to cut through the hype of “Hey, AI is going to take over the world and we’re all going to be living in the world of Terminator in the future.”
Really try to look for practical applications of AI and how you can leverage them yourself in your career to get ahead.