In last week’s blog, we focused on how your organization can stay ahead of the CX innovation curve while remaining true to the core principles of customer-centricity and empathy. In addition to giving our thoughts on the quality of Chat GPT’s response, we outlined four “bold bets” you can make with AI to get more ahead of the innovation curve.
This week, we’ll look at ChatGPT’s answer to how AI can be used to automate and personalize customer interactions, as well as how it can help optimize human-to-human interactions. Like before, we’ll quickly share some thoughts on how well ChatGPT did answering this question, then share our own perspective.
Chat GPT’s Answer (repeated here, verbatim from ChatGPT):
AI can be applied to optimizing human-to-human experiences in a number of ways, beyond just automating and personalizing digital experiences. Here are some examples:
Language processing and sentiment analysis: AI can be used to analyze customer conversations with customer service representatives, either through chat or voice, to identify customer sentiment and provide real-time feedback to the representative. This can help the representative tailor their approach to better meet the customer's needs and emotions.
Behavioral analytics: AI can be used to analyze customer behavior in physical environments, such as in-store or in-branch interactions. This can help identify patterns in customer behavior and preferences, allowing businesses to optimize their physical environments and processes to better meet customer needs.
Predictive analytics: AI can be used to predict customer needs and preferences, even in human-to-human interactions. This can help representatives anticipate customer needs and provide proactive service, enhancing the overall customer experience.
Augmented intelligence: AI can be used to provide representatives with real-time insights and recommendations to help them provide more effective service. This can include recommendations on products or services based on the customer's behavior and preferences, as well as best practices for handling customer interactions.
Collaborative tools: AI can be used to facilitate collaboration between representatives and customers, such as by providing real-time translation or facilitating remote interactions. This can help businesses expand their reach and better serve customers in different regions or with different needs.
In summary, AI can be applied to optimize human-to-human experiences in a variety of ways, providing real-time insights and recommendations to help representatives better meet customer needs and preferences. By leveraging AI in this way, businesses can enhance the overall customer experience, both in digital and physical environments.
ChatGPT’s answer to this question is pretty good. It calls out several ways AI can help optimize human-to-human experiences and consistently drives home the importance of leveraging data-driven insights to help frontline employees focus on ways to connect better with customers. Our primary critique is that ChatGPT could have gone much deeper into AI’s potential to help you understand the emotions that your CX evokes as customers interact with your brand and then drive continuous improvement in your CX design and frontline training accordingly.
Behavioral Vs. Emotional
Before going further, let’s elaborate on the difference between analytics focused on behaviors vs. emotions. Behavioral analytics focuses on what people are doing - whether your customers or frontline employees - but not explicitly on what they are thinking, feeling, and saying. Behavioral analytics can help infer their emotions, for example a shopper with a high dwell time that doesn’t ultimately pick anything up off the shelf and put it in their basket is likely feeling frustrated, but it doesn’t allow you to dive deeper into their specific emotions without further data. Behavioral analytics tends to focus more on operational (O) data that typically comes from your IT systems, but AI is helping to tap insights from additional sources of unstructured data such as video.
Analyzing Customer Emotions
Analysis of emotions requires tapping into experience (X) data that goes beyond what you can capture from structured survey data. AI models can help you dive deeper and more flexibly into your unstructured data, such as voice-to-text call transcripts, social media, messaging, or open ended questions in surveys. It can also include more labor intensive approaches to gather qualitative insights, such as in-depth ethnographic interviews during the customer journey, but these are harder to scale and to apply AI to them (Lou and I share more on how to overcome this via technology-enabled approaches to your customer insights in our Reimagining Insights blog series). Both X and O data offer their own treasure trove of insights about your customers, which can be made accessible by striking the right balance between analysis of behaviors and emotions in your AI architecture.
Understanding your customer's emotions helps you to design experiences that evoke the emotions you want along the customer journey and reinforce your brand strategy. This understanding also allows employees in customer-facing roles (e.g., agents, call center employees, salespeople, etc.) to leverage clues from their interactions and provide more meaningful customer experiences as a result. This is an idea Lou terms “clue consciousness” in his book, Clued In (to see Matt’s review of Clued In, click here). Organizations who train their employees to be “clue conscious” can create a virtuous cycle where more clue conscious employees are better able to engage in the behaviors that you want to evoke the desired emotions for your brand strategy, and you can then leverage conversations among the employees as an ongoing catalyst for continuous improvement.
Bridging the “Brand Canyon”
More emotionally engaging experiences play an integral role in closing the “brand canyon” (the gap between how customers feel about themselves and how they feel about your brand, as discussed by Lou and me in this blog.) To determine gaps between your customer’s feelings about themselves and your brand, consider using an emotional motif. First, pick three to five emotions you want your customers to feel across all interactions, then leverage AI to understand the congruence between those emotions and what your customers really feel.
Getting at emotion does require a well crafted approach, but my partners for this blog, Farlinium and Experience Engineering, have incredible experience helping companies tap into emotional insights. I’m collaborating with Experience Engineering on scalable approaches to Reimagine Insights, and with Farlinium for pragmatic and proven approaches to build out your AI capabilities using the Qualtrics XM Discover platform.
Up Next Week
In next week’s blog, I’ll be writing about leading practices for designing AI-powered interactions. I’ll continue posting weekly for this AI series every Wednesday covering each of the ten questions, collaborating with my business partners for JourneySpark Consulting on these blogs. I’ll also continue to post one of my book reviews for my favorite ten business books every Friday. If you’d like to jump ahead to see all ten reviews now, click here to check them out.
I'm looking forward to continuing the conversation!