The Future of Customer Service: Can Bots Actually Work?

Bots have become an omnipresent part of daily life. From ordering tacos to raising support for charities, they help us live more productively. Beyond their personal applications, chatbots can play an equally helpful role in the realm of customer service (CS). As consumers express increasingly higher expectations for customer support teams, companies are turning to Customer Service Bots to ease the burden on CS professionals, break down organizational silos, and provide targeted data across departments.

For many organizations, the rapid development of chatbot technology sparks plenty of questions. Where do bots fit in a modern support strategy? Are Customer Service Bots here to stay, or are they just another passing tech trend? And how will they change the future of support? We asked the experts.

Looking Back: A Time Before Customer Service Bots

In the past, consumers often had to walk through a complicated series of steps to receive support from a company. They might first have to Google to find a helpline number, navigate through a series of numbered prompts until they’re connected with a support rep, and reconfirm identifying information several times—all before receiving any support.

Providing support was tough for CS agents, too. They faced a daily barrage of customer questions, but according to Agent.ai, up to 80% of all inquiries were low-level and repetitive. Before formally founding Ada Support—a platform that allows companies to build custom Customer Service Bots, CEO Mike Murchison experienced this problem firsthand. “For months, my co-founder and I manually responded to thousands of support emails for six companies. We discovered that virtually every company had the same problem: repetitive questions asked via email, live chat, and phone calls,” he says. “We thought: there has to be a better way to do this.”

How Do Customer Service Bots Really Work?

Many Customer Service Bots use algorithms that access data stored in connected apps. With that data, Customer Service Bots can provide intelligent answers using a type of machine learning known as deep learning. Speech, data, and specific patterns are transmitted through layers of a neural network, which feeds a bot the right information. The greater the amount of data analyzed, the smarter the results. With Natural Language Processing, bots can interact with humans using colloquial, everyday language. Companies can even create custom bots that align with their brand’s unique personality and voice. The end result? Highly personalized interactions that feel almost human.

All this technical complexity is aimed at a simple goal: providing the customer with an accurate answer. So why choose to invest in bots, rather than more human agents?

According to Murchison, Customer Service Bots offer three distinct advantages over human agents. To begin with, bots are available 24/7—even on weekends and holidays. They also offer instant answers, so customers don’t waste time waiting on hold or for a live chat response. And the UX feels comfortable to customers. “The bot interface—a chat window—is familiar,” says Murchison. “It’s like Facebook Messenger; everyone understands how to use it. And elements like videos, GIFs, and emojis enhance the support experience.”

Customer Service Bots also benefit the support team itself. “By using bots to deal with lower-level inquiries, CS teams can spend more time answering complex questions that are more valuable to the business,” Murchison explains.

Moving Forward: To Preemptive Support and Beyond

Gartner predicts that by 2020, customers will manage 85% of their relationship with an enterprise without interacting with a human. As bots become a mainstay of CS, Murchison anticipates they’ll evolve in two key ways. First, they’ll become increasingly personalized. “Many bots today provide helpful answers, but they could be made even more helpful if they had a deeper understanding of who you are,” he explains.

They’ll also be more proactive. “Instead of engaging with a bot when you need help, bots will anticipate that you might need a hand and reach out to you before you realize you needed help in the first place,” he says.

Preemptive support, especially for digital or technology-based products, is already defining the future of CS. By 2018, six billion connected devices will proactively ask for help. A leading cafe chain, for example, integrated Splunk and ServiceNow in order to preempt issues with the self-service iPads in their restaurants. In a world where support is increasingly proactive, it’s easy to imagine bots reaching out to customers before issues even arise.

Why Aren’t More Companies Using Customer Service Bots?

51% of consumers believe businesses should be able to respond to inquiries 24/7, but businesses take ten hours on average to reply to customer messages, according to Oracle. Chatbots offer a cost-effective solution to that problem, so it would seem they’re here to stay. But some companies remain skeptical of bots. Why are these organizations holding out?

Some organizations believe Customer Service Bots just aren’t fit to be the frontline of support. “There’s a misconception that chatbots aren’t good enough to be customer-facing,” Murchison explains. He quickly puts that idea to rest, citing customers’ preference for quick solutions. “In reality, customers are more likely to interface with a bot, because they know they’ll get an instant answer.”

And, of course, there’s the fear that bots will take away human jobs. In Murchison’s experience, however, they’ve done the opposite. Though bots may do most of the heavy lifting when it comes to low-level support, customers will still sometimes require a human touch. “As bot technology becomes more pervasive, there will be a premium on human-driven customer interactions,” he says. The adoption of bots itself has also spurred the creation of new positions. “Our clients have created “bot managers” or “AI managers” who control the bot’s automation processes,” explains Murchison.

RJ Owen, Director of User Experience at Universal Mind, muses that companies aren’t turning to bots because consumers don’t know where to look for them or how to use them. He puts the responsibility on the companies to find ways to bring the user to the bot: “Companies are going to need a lot of messaging around where their chatbot can be reached. [For example] every message sent to a Uniqlo Facebook account should have an automatic response directing the customer to try reaching out through the bot account instead (or the bot should just respond on all of them).”

The larger reason, Murchison believes, is that some companies just don’t care enough. “The businesses that use bots are the businesses that are interested in taking advantage of customer insights,” he says. “Businesses that neglect customer conversations—who don’t feel they’re important—are the last ones to adopt bot technology.”

In Bots We Trust

The general consensus seems to be that Customer Service Bots are definitely here to stay; companies just haven’t figured out best practices yet. As Microsoft’s CEO, Satya Nadella said: “People-to-people conversations, people-to-digital assistants, people-to-bots […] that’s the world you’re going to get to see in the years to come.”

That means the businesses that reject bots will become increasingly less competitive. Their CS reps will continue to waste time putting out fires, rather than preventing them. And because always-on bots are more cost-effective than a nine-to-five human staff, organizations that do adopt them can reinvest the savings into improving their products and customer experience—making them even more competitive.

To ease the transition to a bot-inclusive CS strategy, Murchison suggests taking a long-term view. “It’s really about recognizing that customer conversations are a massively underleveraged but incredibly valuable resource,” he says. “When you can interact with your customers instantly, you can learn more about them than ever before.”

Before there were Customer Service Bots, there were Self-Service Help Desks – is yours working the way it should be? Evolve your Help Desk >