There's no end to the hype around customer experience and artificial intelligence these days. Between the Internet of Things (IOT), machine learning, chatbots, virtual assistants, Watson, Einstein, Alexa, and Siri, everyone's pitching a customer experience paradise where customers simply think of what they want and drones magically appear on their doorsteps … almost.
The reality is that customer service, customer engagement, and the adoption of emerging technologies in the contact center, is still a work in progress for most organizations. For most companies, the prospect of adding new technologies can be a bit overwhelming. We see many customers taking a wait-and-see approach to many of the newest technologies. Those on the cutting edge hope they do more to delight than offend customers.
The good news is that technologies that can have a real impact on the bottom line by increasing agent productivity and case resolution do exist. In fact, our research finds that adding intelligence to customer service can increase agent productivity by an average of 25 percent and manager productivity by an average of 21 percent. So, to achieve those kind of benefits without disrupting customer contact channels that are working, where do you start? Whether you're an early adopter or a wait-and-see skeptic, here are a few steps to take to drive both bottom-line and top-line results.
Step 1. Bring bots to case management.
Remember Eliza, the virtual therapist? Chat bots have come a long way since then, and customer service organizations can leverage them to resolve simple customer requests (close cases), gather information about customer needs (reduce case resolution time), and supply relevant information to agents working on cases (increase agent productivity). It's important to note that bots are no longer confined to the texting or e-mail channel: as consumers get more used to Alexa and Siri, they will become more comfortable interacting with bots in the voice channel. Another important thing to remember is that with bots on any channel a customer should always have an easy way to get to a human. American Airlines, for example, employs voice bots even for initial contact from premier customers, but at any time a customer can say "agent" and be directed to a live person.
In the future, look for your customers to start using their own bots to initiate and escalate cases, creating more of a blurred line between bots and real agents.
Step 2. Invite learning to the knowledge library.
In our analysis of customer service maturity, we often find companies take a big leap forward when they focus on curating their knowledge library to provide more self-service options for customers (and more effective and case-relevant information to agents). The benefit of bringing machine learning and artificial intelligence to the table is that, unlike sales or marketing data, service data is much more discrete and complete in nature, and cases follow a common pattern. This enables companies to apply AI capabilities like Salesforce.com's Einstein or IBM's Watson, for example, to their existing knowledge libraries and case data to catch patterns and analyze them for recommendations. These technologies learn over time about what knowledge is most valuable in resolving cases and what may be missing.
In the future, look to static knowledge libraries to become a thing of the past. Customers will be able to surf (or have their bot surf) the information they need, and self-learning knowledge libraries will be dynamic, with little or no human intervention.
Step 3. Get predictive with management.
Bringing intelligence to existing service data is a great way to help managers better understand why customers are calling, which agents are doing the best job or need remediation, and which issues or events might be driving spikes in service calls that need to be addressed with more proactive service. The good news is that improvements in text analytics can bring not just real-time but predictive recommendations to service managers, enabling them to get ahead of issues before they become problems.
In the future, look to predictive capabilities to eliminate much of the tasks of contact center managers today, as they can intelligently route cases to the appropriate agents, drive training in context for agents with skills gaps, and take a completely data-driven approach to agent management.
Step 4. Text with your customers.
As more consumers turn to texting instead of calling, the ability to support multiple chat applications and SMS will be a requirement. If you haven't enabled SMS service yet, now is the time to start. The good news is that SMS resolutions cost much less than agent calls, and bots (yes, they're here to) can help you automate much of the communications.
Step 5. Get real with agents.
The human agent is not going away any time soon, but the agent is also being forced to interact with customers who are more tech-savvy and have higher expectations. In analyzing the experience of customer service agents, we have found that traditional agent training and motivation often misses simple, human ways to keep agents performing and improving. It won't cost you a dime to have a real conversation with agents to determine what's most frustrating for them, and you'll often find that simple training on how to use the productivity tools already within their agent desktop (such as hotkeys, case templates, and the like) can keep them humming along with the bots.
Rebecca Wettemann is vice president of Nucleus Research.