How to seize the potential of generative AI in digital CX, according to leaders

Much has been made about a future unlocked by generative AI (GenAI) and the changes that the technology will bring. The time for waiting and wondering has come and gone, however. It's now time for brands to look at generative AI in the present tense — to experiment, to iterate and to scale.

Customer experience (CX) leaders are rising to the challenge and dedicating their resources to implement AI solutions. According to Deloitte's report, Trends & AI in the Contact Center, 81% of contact center executives are investing in AI for agent-enabling technology. The same report indicates that 81% have invested in voice and text analytics. These brands are not waiting for what's next; they're building it. Still, there is a need to approach this inflection point thoughtfully.

Applying generative AI in your CX operation should not be a haphazard or slapdash exercise — the trust and experience of your customers and employees is at stake. Fortunately, there are a number of best practices offered up by TELUS Digital leaders that will help you seize the potential of generative AI while mitigating risk. Let's take a closer look.

Best practices for generative AI in digital CX

Whether you have been proactive in bringing GenAI to your customer experience, or you're just starting now, keep these tenets in mind as you progress.

Operationalize your data to set your CX apart

If you're out to deliver a differentiated customer experience, you need to lean into your understanding of your customers, employees and product. And to do this, you've got to operationalize your data.

Your data is what really matters, according to Michael Ringman, chief information officer at TELUS Digital. In a recent Diginomica interview, Ringman underlines the importance of "understanding the knowledge of your data and leveraging that to its fullest," rather than taking a generic GenAI product from the market and shoehorning it into your CX operations. For example, if you're using a generic solution to try to predict customer retention risks, but that solution doesn't understand the nuances of your business or customers, you may not be doing your CX any favors. Likewise, Ringman insists you can't simply swap in an off-the-shelf GenAI chatbot "and expect that to take away your call volume," before emphasizing, "It's just not going to work like that."

For many, effectively putting your data to use is easier said than done. As it turns out, accessing and understanding untapped data is often where brands need help. Ringman explains that the data brands are looking at today tends to be limited to traditional contact center metrics, whereas they should be analyzing the data that will help them to understand where customers "want to be met." Partnering with a CX provider that has expertise in data analytics and artificial intelligence can make a big difference, helping brands to leverage their own data to deliver bespoke experiences.

Reduce customer and agent effort with bots

Once you've operationalized your data, you have the perfect opportunity to introduce GenAI-powered chatbots that can make finding answers easier for customers and agents alike.

In an article for Forbes, Brian Hannon, chief growth officer at TELUS Digital, states that generative AI "has the ability to better personalize and contextualize customer-bot interactions," before adding that "it won't be long before GenAI is automating a greater number of CX needs and inquiries, thereby enabling faster and more efficient customer interactions." The latest GenAI chatbots are exceedingly capable of understanding complex prompts, and when they are trained on your own company data, they can field many of your customers' questions. That gets them the answers they need quickly, and makes it so that agents are freed up to manage complex inquiries that require human empathy and attention.

On the subject of your customer care team, Hannon explains that "GenAI can greatly enhance support tools used by contact center agents to act as co-pilots, working behind the scenes and alongside agents to complete tasks such as quickly summarizing large volumes of data to tailor recommendations and responses to questions." In this way, bots can help surface information for agents so that they can answer customer questions, but that's not all. Bots can reduce agent effort further when they automate time-consuming tasks, such as entering case information into CRM systems.

Combine conversational AI and GenAI to create voice-first experiences

"Exploration and early adoption will be critical in distinguishing the winners in today's highly competitive AI implementation battles. However, the primary differentiator for companies to succeed in this crowded space is not simply generative AI alone. Instead, it's GenAI paired with voice technology," argues Tobias Dengel, president of TELUS Digital Solutions & WillowTree, a TELUS Digital Company, in a Fast Company article.

Dengel, who authored a book on voice-first experiences titled The Sound of the Future: The Coming Age of Voice Technology, makes a compelling case for how voice technology should be used to usher in the next generation of search, navigation, data analysis, translation services and more. As Dengel states in his Fast Company article, the key is to integrate generative AI algorithms with multimodal voice technology. Generally speaking, "This means combining the most immediate input (voice) and output (screens) with the natural language capabilities of GenAI," says Dengel. For example, because people can speak faster than they can type and read faster than they can comprehend a spoken response, multimodal voice technology enables a customer to effortlessly voice a question to a chatbot and have the response come back to them on their screens.

More than a mere "UX enhancement," Dengel urges that voice-first experiences "will quickly become a requirement in next-generation, AI-enabled software."

Hone your AI-powered experiences with a human-in-the-loop approach

By now, those familiar with generative AI's potential in CX will also be aware of the associated risks. Hallucinations and model bias may raise questions, but they do not have to go unanswered. To ensure generative AI implementations are performing as they should, brands must implement guardrails and take a human-in-the-loop approach.

Ringman highlights the importance of guardrails within generative models in a recent Forbes article. "These guardrails act as constraints or rules that guide the AI's output generation process, ensuring the content remains within acceptable boundaries. By defining limits on the generated content, such as adhering to known facts or respecting ethical considerations, we can significantly reduce the likelihood of hallucinations and produce more accurate and reliable results." To offer an example, if you're a retailer with a customer-facing chatbot powered by GenAI, you want your bot to be able to handle questions about your products and return policies, not history or political theory. Guardrails can be used to limit the scope of topics that your chatbot will comment on, and the data sources that it learns from.

Beyond guardrail development, humans play an important role in validating and monitoring GenAI outputs for bias and hallucinations. "In the context of GenAI hallucinations, having humans in the loop means incorporating humans who can review and assess the generated content for accuracy and coherence. They provide feedback, identify potential hallucinatory or misleading outputs and make necessary corrections or adjustments to ensure the generated content aligns with objective reality," explains Ringman.

In an article for Fast Company, Jeff Puritt, president and chief executive officer at TELUS Digital, takes the idea of human-in-the-loop a step further. Puritt explains, "In my view, the key lies in a paradigm shift beyond keeping humans in the loop to embracing a 'humanity-in-the-loop' model. This holistic approach recognizes that we — people and technology alike — are part of a complex ecosystem that requires uniquely human expertise and thoughtful governance."

Building trust in generative AI

Brands are eager to reap the benefits of generative AI (GenAI) while limiting potential risks. Join Steve Nemzer, director of AI growth and innovation for TELUS Digital (formerly TELUS International), as he shares best practices for leveraging GenAI without compromising your organization’s goodwill.

Watch the video

Design, build and deliver AI-fueled intelligent experiences

While generative AI's potential in customer experience seems limitless, it can be difficult to progress past speculation. To apply the technology within your organization, you need to know how it can integrate with your existing way of doing things and the best practices to follow to increase your chances of success. For CX leaders looking to make forward strides with generative AI, a knowledgeable partner can help.

If you're looking to seize the potential of generative AI in CX, contact our team of experts today.