Providing an exceptional customer experience has always been the goal, but in today's world of social networks, mobile apps, messaging, phone (yes, voice does still exist), and chat bots, there are myriad ways for customers to interact with brands, and it can be overwhelming. Getting the channel correct is one thing, but being smart about how you communicate is a whole different matter.The best service organizations learn from their service successes and failures.
I personally remember setting up a new video game system I bought my son for Christmas and, for some reason, waiting hours to download and update the software. I only found out why it was taking so long on Twitter (where many of us go these days). I originally thought it was a problem on my end, but, as it turns out, it was a system outage on the server side. It would have been nice to have been texted or emailed to know there was a problem. The manufacturer's brand was getting tarnished on Twitter as anger grew during the outage.
I contrast that experience with being upgraded by JetBlue without even asking. I was surprised with a new boarding pass, going to the gate agent and getting a Mint seat.
As customers, we don't forget good experiences, and we never forget bad ones. The challenge for companies is to learn and improve experiences.
There are many complexities to service centers related to channels. The term is often called omnichannel. The term itself represents the fact that customers want to receive information on any channel of their choosing when and where they want it. They expect service to be personalized, as if you were servicing them on a one-to-one basis. Since we can't practically ask each and every customer their needs, or manually relate one customer experience to another, we need to scale the institutional knowledge we have obtained over time. Artificial Intelligence (AI) makes that possible, which is why it's one of the hottest technologies that service centers are adopting to improve customer satisfaction. Let's look as some specific use cases.
Smarter Supervisors
The goal of a service supervisor is to ensure agents are productive and, most important, are providing high levels of customer satisfaction to customers. The core metrics on which supervisors focus are customer wait (queue) time, time to resolution, survey feedback, and agent availability. The nirvana state for any supervisor is not to wait for a customer to call about an issue but to proactively reach out the minute an issue is known. This seems relatively straightforward but is actually more difficult in practice. The major stumbling block is monitoring and understanding data as it become available.
I recently got my son a miniature drone. After only one flight, one of the rotors stopped working. Other friends from different parts of the country who bought the same drone were not having the same problem. So what do I do? I go to the store to get another one. Sure enough, same problem again. So now I am annoyed and call the manufacturer. I asked the agent if he could see if the company had gotten similar complaints. The agent says that it has (why did I have to tell him!), but it seems mostly from the Northeast (where I live). Ironically, my son registered the drone, so the company had my information and knew there was a problem. But because the company had no tools in place to discover the issues through their own data, the call center just continued to get flooded with calls (translation: long queues, poor customer satisfaction). Though this is a simple case, it's easy to extrapolate how this issue could come up in any business. Ideally, as this information was coming in, there would be AI acting on the data constantly to detect potential patterns that could notify a supervisor so a proper escalation process and remedy could take place. AI would make supervisors smarter and empowered to make more specific targeted recommendations to improve customer experiences.
Smarter Case Responses
It is often misunderstood that AI will replace people. In customer service, AI will make agents smarter by learning how previous cases were resolved. Too many times, a service agent will see a case for the first time, even though it might have been logged by other agents countless times before. After receiving a new case, the agent often puts a customer on hold and frantically searches to get a solution. Here is the first place AI can help. With AI, the agent would be served information about similar cases that were previously solved, including tagged information such as knowledge articles and digital content. In essence the agent is now leveraging the experience of others who have solved the same issue before because the case resolution is smarter. Even in a case where no one has solved a similar problem before, AI can find information related to the case by looking at similar examples. AI can also route cases very quickly to the most appropriate agent based on skill set and level of expertise. AI-based case handling simply makes routing and answering cases more efficient and effective.
AI will not be replacing jobs but enhancing them. Think of the example where a consumer calls a service center because his smart TV has a glitch when playing movies from the Internet. There are many potential sources of failure, ranging from the TV itself, to the wireless Internet hub, to possible bandwidth issues. A consumer is most likely to start with the last point of failure, which is the TV manufacturer itself. However, regardless of where he starts, information needs to be collected on all the possible points of failure. The first thing the company should do is get basic information, possibly by text. This interaction could be a back and forth with a messaging bot. In the meantime, AI could identify the best agent to call back to help the consumer. Once this is determined, a text could be sent to the consumer asking if he would like to further discuss the problem. The value from this interaction is that even if the problem is not with the TV, the information will help for the next case and enables the service agent to provide value to the total solution the consumer has, not just with the equipment he called about. The next time this happens, the learnings from this case resolution will make the agent even smarter because the case is smarter.
Smarter Mobile Service Employees
We've all lived it. Waiting for a field service technician to show up at our house, only to find out he did not have the part or didn't know the model of product he was there to service. We then end up back in queue for the service technician to make another visit. There are several parts of this process where AI can make mobile services employees smarter. The first is creating an optimized route and schedule. This would include estimated work time based on other jobs , geography, and parts availability. The other crucial AI component is to get as much information about the case before the field technician is even scheduled. This could include digital image diagnosis or sound analysis remotely as the case is being submitted.
Let's take the dreaded burner having a problem in the basement. In this example, the consumer will be asked to take an image of the actual burner (to recognize make and model), an image of the filter (to check to see if it is worn), and if possible, a recording of the sound as the burner heats up. With these three pieces of information, the case can be logged and handled. Now let's say it was a filter issue because the AI engine analyzed the digital image and realized that the filter has not been changed in a year. The customer can get on-the-spot advice to change the filter. Case closed, no technician required. However, the case might go further than that, particularly if the filter looks OK but the sound the burner is making dictates another issue that will require parts to be replaced. Now the technician can be contacted, the parts set aside, and the appointment scheduled. When the technician arrives, he already hasd the solution and is not trying to find the problem.
The common thread in all three of our examples is that AI is used to make service professionals smarter to improve customer experiences. By providing supervisors, agents, and mobile employees with AI-fueled information we can truly transform customer service processes.
Robert DeSisto is chief value officer at Salesforce.com. Previously, he was a vice president and distinguished analyst at Gartner , where he was responsible for managing the software-as-a-service (SaaS) research agenda, CRM sales agenda, and business application research community.