Customer service teams deal with higher expectations than ever before; customers want near-immediate solutions to their problems. Staffing customer service teams around-the-clock can be costly and tedious, so many organizations are turning to incorporate more technology into the customer service experiences they offer. As a product manager that works to use AI to make customer service even better, Sophia Xing often hears that many customer service professionals feel like new technologies are their enemy. Can a chatbot replace people entirely? Will the most sophisticated artificial intelligence technology make human support teams obsolete? Sophia Xing says that's not the case.
So what makes Sophia an expert on the matter? She's a data enthusiast at her core, but she has spent her career applying her technical knowledge to real-world human challenges. She studied at Princeton University, receiving a BA in Economics, Statistics and Machine Learning, and Applications of Computing. After getting four years of work experience at Goldman Sachs working on quantitative investment strategies and using AI to build alternative investment portfolios, Sophia went on to get her M.B.A. at MIT Sloan School of Management. There, she honed in on natural language processing, mobile and sensor computing, and computer vision.
Her education focused on cutting-edge technology, but her career quickly turned into applying technology to solve problems and develop products. Product management, the field Sophia entered after MIT, is filled with professionals who can merge technical know-how with applied problem-solving and strategic skills. At Google, Sophia worked on the development of features in System Trace, an advanced tool in Android Studio. Then, she joined Bodeswell as the first product manager at the organization. She launched new features including back-end ranking algorithms and front-end UI customizations.
Eventually, Sophia made her way to Forethought. There, she's a Lead Product Manager and spearheads a team of engineers, designers, and researchers who are using machine learning and AI to solve customer service problems. Applying machine learning in customer experience interactions has allowed Sophia to see the tension that exists between customer service providers and the tools that can help revolutionize their capabilities.
While people often feel that AI and other technologies are going to replace them, Sophia has seen that the best approach is for AI customer service agents and human agents to work as a team. AI agents can help automate a lot of tedious tasks and basic inquiries, freeing up time for human agents to focus on more complex customer inquiries that require their intervention. Without a seamless partnership with technology, human agents are setting themselves back and organizations are failing to provide the most timely customer experience to their customers. But, on the other hand, taking humans completely out of the process would mean that inquiries truly requiring human discretion would not receive the needed attention. The future of customer service straddles the line between human support and AI automated agents.
In order for human and AI agents to collaborate seamlessly, a few capabilities need to be established. Firstly, the AI agent needs to be able to recognize what the customer is asking and establish a judgment of whether it is a feasible question to auto-resolve. If not, then the AI agent needs to be able to direct to the human agent right away, preferably at the beginning of the conversation in order to preserve a low level of customer effort required. Generally, an AI agent is good at tasks in information retrieval (finding the most relevant article), logic-based task automation (refund orders when delay is over a certain time period), and large-scale information gathering. On the flipside, humans are usually better at exercising discretion when offering similar solutions, serving as a sounding board to make sure customers feel that their feedback is heard by a representative from the company, as well as processing one-off cases that may not be detailed in company customer policies. In the long term, the overall system needs to be able to recalibrate - to recognize what questions that it thought could be answered by an AI agent are actually proving false, and vice versa, which questions flagged for human agents to answer are actually something that can be handled by an AI agent.
There's virtually no customer support team that isn't overloaded with requests and struggling to keep up. Sophia recommends that these teams see AI as a tool to make their roles more manageable instead of a new technology that is going to be a detriment to their careers. Helping humans will always require humans, but technology can help optimize processes and make customers and agents happier with the end results. As a product manager, Sophia works diligently to help her clients understand this dynamic and implement the right mix of tools and new processes to lead to more customer success stories.
