The End of the Age of Frustration: Artificial Intelligence Will Help Create Better Customer Service Agents


Share on LinkedIn

You’re stuck at the airport because bad weather has thrown the flight boards into chaos. There’s a line at all the desks, so you call the airline. All agents are busy with other people who have the same problem. By the time someone calls you back, you’re in a foul mood and impatient with the agent who has to look up your information and so on. Everything moves slower than it needs to.

Let’s introduce artificial intelligence into this scenario. Now you can get online, where a chatbot guides you through possible solutions. It’ll have your customer information and be able to find you a new flight and update you on the weather through push notifications.

That’s the vision for the future of service: intuitive, self-guided AI systems that will take care of the most common customer issues. Does this make the service agent obsolete? Far from it. While AI handles the nuts and bolts, it facilitates the evolution of the agent’s role — AI takes care of the IQ while your agents focus on the EQ.

Serving the Service Agent

The world of customer service is demanding. Service agents are expected to find the right answer instantly, to know who customers are and what they purchased, and, occasionally, to be a therapist. AI will help with the first two, but the last is a uniquely human task: to empathize, to understand what you, as a customer, have been through, and figure out how to make you feel better about it. That’s EQ, the measure of emotional intelligence, and it’s an essential quality in providing amazing service that keeps your customers happy, even when they’re having issues.

This is where AI can really help agents provide the level of service customers expect.

For simple, frequently asked questions, like “How do I reset my password?” “How do I start a tax return?” or “How do I change a flight?” companies can use AI to answer the questions with chatbots.

Companies worry about how customers feel when a machine answers the phone. But research has shown that most customers actually love the technology, as long as it’s being used in the right place in their journey — and they know that it’s a bot. Customers like that bots offer quick answers. They’re even willing to give a bot key information if they know it will get them faster and better service with a human agent who won’t ask them the same questions. Break that implicit promise, and you could lose the customer. Keep that promise, and you will end up with a happy customer for life.

Developing An Agent’s EQ

AI will be more than a simple bot-based Q&A system. It can figure out what the customer’s issue is based on previous purchases, find an answer by looking at similar cases or route the call to the right agent based on the expertise needed. Basically, AI will handle low-level queries directly and do all the searching through help articles and even external information sources to deliver answers to agents. So then: what does the agent do?

Agents get to be human. Freed from tedious information hunting, they can take the time to understand the customer and the issue. They can empathize. They can use humor to build rapport. And they can apply human intelligence to augment what they’re getting from AI to make sure that the customer is getting the right answer. For valued customers who’ve had multiple issues, the agent can offer a discount, refund, or replacement product.

Customers prefer solving their own problems whenever possible. Eighty-one percent of them will take care of things themselves before reaching out to a live representative. So the issues agents face are more complicated than ever before — and the customers they’re dealing with are more frustrated since they’ve already cycled through the self-serve solutions. This added pressure means that customer service will be set apart by those organizations where agents can connect with customers, process customer needs, and provide the right tone, the right amount of empathy, and the right understanding. The new service agent must be able to provide complete customer support with a generous helping of emotional support.

Better Machines for Better Agents

Customer expectations are high. The pace of customer satisfaction is set by the big players like Amazon and Google, companies that can provide delivery information and products nearly instantly. The slow, inefficient nature of traditional service looks more and more antiquated in this environment, and customers know it. They don’t want to wait. They don’t want to give the same information over and over. There’s only so much hold music a customer will listen to before switching to a more engaged company. Accenture reported in 2016 that 52% of customers switched providers due to poor service, a cost of $1.2 trillion.

Service agents’ expectations are also high. They see Amazon and Facebook serving up relevant offers based on their past purchases (or those of their friends’), and they know these technologies can make them smarter and faster at their jobs. Service agents will follow customers out the door if the company doesn’t modernize.

AI-powered service is showing up in all sorts of companies. In fact, eight of 10 companies plan to implement AI into customer service by 2020. This is the new normal, and for the sake of all of us who have been caught in airports, forgot a password, or just want some of some lost hours back, it’s about time. The need for EQ extends beyond service — all the way to the top of any modern company.

Sarah Patterson
Sarah Patterson is responsible for the messaging, positioning and go-to-market strategy for Service Cloud, the world's #1 service solution. Service Cloud empowers companies with a connected platform to deliver smarter, faster and personalized customer experiences. Sarah brings more than 15 years of experience in IT consulting, research, and marketing to Salesforce and has held a variety of positions at McKinsey & Co., Broadview International and the Initiative for a Competitive Inner City.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here