Exclusive Q&A: Sweetwater chief sales officer discusses ChatGPT

The new ChatGPT artificial intelligence (AI) platform may not be the right tool for every customer service interaction.

Chain Store Age sat down with Jeff Radke, chief sales officer of Indiana-based online music technology and instrument retailer Sweetwater, to discuss customer service and how AI technologies like ChatGPT can enhance, but not drive, retailers’ customer engagement strategies.

ChatGPT is a new AI model from research and deployment company Open AI. Using 570 GB of publicly available data on the Internet, ChatGPT interacts with users in a conversational style that mimics human interaction and uses machine learning (ML) to continually refine and improve its responses.

[Read more: ChatGPT is coming – what it means for your enterprise]

What are the drawbacks of using ChatGPT for customer service? 

ChatGPT can’t maintain the level of service that loyal customers come to expect after experiencing competent, caring and empathetic human interaction. When companies communicate with a customer, it adds to or subtracts from credibility as an employee and as a company.

That said, ChatGPT is only as good as the data it’s trained on. If it’s not consistently updated with fresh data, it could spread misinformation and cause customers frustration, leaving the customer with a negative perception of the company as it creates a barrier to resolving their issue.

Furthermore, relating to a customer goes far beyond just answering questions and finding solutions. It requires empathy and compassion. Detecting when a customer is confused, frustrated, rushed or excited is critical to understanding and serving them well.

With a company like Sweetwater, the complexity of the products we sell; and the need to truly understand the customer’s needs, what they are familiar with, and what their goals are, chatbots are just too cumbersome. It’s so much faster to simply speak person-to-person.

Are AI-based chatbots a viable customer service solution in general?

ChatGPT can be a viable customer service solution in some situations. The technology can generate some surprisingly human-like and professional communication, so it could be utilized to handle very basic customer service-oriented questions.

In addition, leveraging this technology to supplement certain tasks can be incredibly powerful. For example, if ChatGPT were used to generate a first draft of a response, and then an employee edited it to add a human touch, it could significantly increase efficiency while still allowing communication at a personal level.

Still, nothing replaces the value of speaking directly with another person to obtain maximum quality and depth of communication.

What do you see as the future of ChatGPT in retail?

I believe that many retail companies will use ChatGPT and generative AI tech in the future. Retailers may use it to handle very basic customer service-oriented questions, if their products aren’t too complex or the situation too nuanced.

But it’s too early to understand where all of this will go. I’m excited about any technology that can help us to better serve our customers, so it will be interesting to see how products like ChatGPT evolve and are utilized in the future.

How can retailers ensure positive customer engagement?

Technology is great, but people matter, so utilizing technology solutions to allow people to work more effectively and efficiently should always be the goal. Simply put, engagement means caring about each and every single customer experience and striving for it to be extraordinary.

It’s easy for a company to look at a Net Promoter Score and feel good about customer service because they scored well overall compared to the competition or other norms.

But about every customer they failed? Is there really an acceptable percentage of dissatisfied customers that a company just gives up on and allows to suffer a negative experience?

What’s been critically important for Sweetwater is to never allow a customer to be a number. If 999 out of 1,000 customer interactions went well, we are maniacal about digging into that one failure and will go to great lengths to solve the problem and fix the customer, which are two very different things.

If the customer isn’t happy, solving the initial problem is meaningless. In the process, we often learn things that will help us improve our overall level of customer service.