The reality of AI in customer service
When we talk about memorable customer experiences, we tend to value what’s human. We celebrate the human touch in a conversation. We take pride in the personable way an agent facilitated a fix or solution. We strive for personalized interactions that wow customers. Yet customer service, like every business function, is being transformed by automation and artificial intelligence. In fact, Gartner predicts that as many as 85% of customer interactions will happen without a human by 2020
Will artificial intelligence make human agents obsolete? Or will this technology augment an agent’s efficacy, helping improve brand consistency and customer satisfaction? We explore the full potential of AI in customer service and what it means for your brand.
WHAT AI LOOKS LIKE IN PRACTICE
We’re living in a world where a computer can beat human players at Jeopardy and improve cancer diagnoses. AI-enabled customer service is well-within the realm of possibility. But what does that look like in practice? The applications of AI vary, but these are some of the most common instances in customer service:
Human agents are still irreplaceable as conversationalists. Yet AI programs like chat bots are now skilled at simulating the basics of conversation – giving preset answers to common questions. Chat bots can help people schedule appointments, track flight changes, answer common customer care questions, and even reroute customers to live agents (after answering some qualifying questions first). Smart machines take charge of the simple requests you receive in bulk so human agents can tackle the complex ones.
Tech support agents are most productive when they’re immersed in complex fixes. The work of assigning and closing tickets slows momentum and distracts agents from what they do best. Today’s AI tools help streamline tech support solutions. Rather than taking a detour from problem solving, agents can rely on triggers within AI programs to automatically generate tickets or close them when the right criteria are met. By monitoring the content of interactions across channels, these tools can keep support processes on track while empowering agents to focus on the tricky problems at hand.
Smart support suggestions
Customer care and tech support agents need to be walking, talking wikis for your brand and/or technical knowledge. Yet even the best agent can benefit from a second opinion. AI programs can now provide agents with smart suggestions, helping them to spend less time deliberating over the right fix. This improves customer satisfaction and first contact resolution.
Customer churn is the bane of any business. Though some churn is unavoidable, many customers are retainable – if you take the right actions. To that end, AI is already delivering results.
Sprint offers a great success story. They had one of the worst churn rates on the market and wanted to make a change. Using an AI-powered algorithm to identify customers at risk of attrition, Sprint took proactive steps to provide those consumers with offers that retained their business. Thanks to this tool, they were able to reduce churn by 10% – and increase transactional NPS by 40%!
FINDING YOUR BALANCE BETWEEN AGENTS AND AI
With all of the above examples, it’s easy to see why some people imagine customer service of the future as more machine than man. However, that way of thinking overestimates the capabilities of AI. In all of the above examples, there is a distinct “if/then” logic at work. AI tools don’t grasp the meaning behind customer requests or even their own responses – they just fulfill conditions in preprogrammed ways.
Moreover, AI lacks the soft skills that define remarkable customer experiences. Machines can mimic subject matter expertise. They can’t mirror the critical thinking, creativity, empathy, and complex communication skills that incredible human agents have in abundance. And it’s that criteria that separates AI and human tasks.
Here are a few examples where the human touch is still king:
Your customer just had a terrible air travel experience. They’re angry and want to feel heard. A canned response from an AI chat bot will feel hollow (and might make the situation worse). Yet a human agent can provide empathy and communicate in a way that validates the customer’s frustration as he or she creates a fix.
Your enterprise client is having difficulty using your technical system. The resolution requires more than a few steps to complete and most of their end users do not have a technical mindset. A human technical support specialist can pick up context clues from the interaction to determine the best step-by-step guidance to give each specific user.
GETTING AI CUSTOMER SERVICE TO WORK FOR YOU
Even in instances that require more of human interaction, artificial intelligence has a part to play. According to ZDNet, customer service agents are better utilized with AI at their disposal. Without access to artificial intelligence platforms, agents spend 49% of their time on complex tasks, but with those platforms, they increase that output to 66%. That means more customers receive a true personalized experience.
Like any tool, AI platforms are only as good as the foundation supporting them. If a chat bot does its job and forwards a customer to an appropriate agent, there needs to be people, processes, and other technologies in place to engage and satisfy that customer. Otherwise, AI is an expensive link in a broken chain.