Generative artificial intelligence (AI) is revolutionizing many aspects of business operations, particularly in customer service. The use of bots to handle high-frequency, low-complexity tasks is well-established. Generative AI is impacting the role of customer service agents, challenging the belief that AI might make these roles redundant. Most customer interactions today begin digitally, with AI-powered bots or intelligent virtual assistants (IVAs), accounting for around 85 percent of these interactions. With the advent of generative AI, bots can handle even more complex transactions.
Despite these advancements, there are situations where generative AI falls short, particularly when customers can’t articulate their problems effectively. This underscores the importance of human involvement in customer service, especially when addressing complicated issues and resolving problems. Paradoxically, AI is making customer service agents both less and more important.
ZK Research recently hosted a webinar with NICE, a provider of enterprise cloud contact center solutions, to discuss the impact that generative AI has had on IVAs and AI-supported agent interactions and how this can create new experiences that weren’t available in the past. A summary of the discussion is below.
The Impact of AI on Customer Experience
Customer experience is now the top brand differentiator, overriding other factors such as price and product quality. Two-thirds of millennials have shifted brand loyalties due to a single bad experience, according to ZK Research data. Currently, 90 percent of companies compete based on customer experience, a significant increase from 28 percent five years ago. A top priority for organizations with digital transformation initiatives is improving the customer experience.
Shortly, most company or customer interactions are likely to be influenced by AI. 78 percent of organizations plan to invest in AI to enhance the customer experience by the end of 2023. Companies that provide a superior experience continue to be patronized by customers (i.e., strengthened brand loyalty). On the other hand, companies that don’t offer a good experience struggle to keep customers.
Generative AI facilitates rapid personalization of customer interactions thanks to its ability to analyze vast amounts of data. Natural language processing (NLP) tools driven by generative AI, such as OpenAI’s ChatGPT, are democratizing AI beyond the realm of data scientists and specialized companies. Such tools have become widely available to everyone removing most barriers to companies at least trialing the technology.
Large Language Models and Evolution of IVAs
IVAs have had a mixed reception, ranging from skepticism to outright dislike, mainly because of their limited intelligence without AI. An IVA is another form of interactive voice response (IVR), which has been more frustrating than helpful for customers, despite its role in customer service. One of the main problems has been the lack of integration between systems, where agents didn’t pick up information input into an IVR. The same was true with IVAs.
With the implementation of large language models (LLMs), IVAs have become more conversational. Instead of using specific phrases, users can now communicate using natural language. Some LLM-powered IVAs have improved to the point where they’re preferred over human agents in certain scenarios due to their speed and no wait times.
Applying generative AI to IVAs has made them smarter, more accurate, and more capable, leading to a significantly improved customer experience. Generative AI-supported IVAs provide a vastly better experience. People can speak in their voice and be understood without molding the speech pattern to the needs of the software.
Shifting Metrics for Contact Center Performance
In the age of AI, traditional performance metrics such as average handle time and first call resolution are becoming increasingly irrelevant. AI and advanced routing systems have enabled contact centers to handle complex transactions like mortgage processing in a single call, enhancing customer satisfaction and potential revenue generation. There is a shift from efficiency metrics to outcome-based ones, where contact centers are becoming profit centers that drive revenue through targeted promotions.
Moreover, the customer journey is now a key focus for contact centers. The new performance metric ensures that customers who need human intervention reach an agent quickly while others are served efficiently through generative AI. This represents a significant change from high-frequency, low-complexity interactions to quality, high-engagement interactions.
Potential Concerns with Generative AI Technology
The transformative potential of generative AI is undeniable—it’s poised to change society and business practices in ways we can’t yet fully comprehend. But, like any technology, generative AI has its pros and cons. It can also fall into the wrong hands and be used in cyberattacks or phishing scams. Other concerns involve potential bias and discrimination from AI algorithms trained on insufficiently diverse datasets, as well as overreliance on AI.
Businesses leveraging generative AI technologies must be aware of potential concerns, such as data quality and consistency, regulatory and compliance issues, and implications for customer trust. For instance, using ChatGPT for making stock recommendations could raise regulatory concerns. Therefore, companies should be transparent about how they’re using generative AI, or they risk losing customer trust.
The Strategic Approach to Generative AI Adoption
Companies looking to adopt generative AI should take a strategic approach that involves identifying the top business use cases. In terms of specific use cases like understanding the nuances in customer queries—particularly in areas like property law—generic AI might not be sufficient. This is where “bespoke AI” comes in, designed and trained to handle more complex queries. Therefore, companies might have to create generative AI models using unique datasets that cover specific needs.
Human oversight remains essential in the application of AI. Just as autopilot in cars requires human supervision, so should AI tools, especially in critical applications like healthcare and finance. All companies should have policies for implementing technologies, such as ChatGPT, effectively and ethically.
Businesses should be asking their vendors how they intend to integrate generative AI and assessing their own data to determine if they are prepared to leverage it. Despite being in the early stages, there will be many new opportunities and challenges ahead, as we move to an exciting future with generative AI at the forefront.
