The Third Wave of CX Leverages GenAI and Human Expertise

Digitally Irresistible

A Veteran in Customer Experience Innovation

We welcome Mark Honeycutt to this week’s episode of the Digitally Irresistible podcast. Mark has enjoyed a distinguished career, marked by significant contributions to the fields of technology, retail, and service. With over 25 years of experience, he has built a solid reputation for creating exceptional customer experiences at iconic brands such as Amazon and Microsoft.

Mark has been involved in three major waves of customer care evolution. In the late 1990s and early 2000s, he was at the forefront of leveraging third-party companies, known as business process outsourcing (BPO) providers, to deliver customer experience solutions. Following this, he played a pivotal role during the offshoring wave, where customer care operations were relocated to different parts of the world to optimize costs and efficiency.

In this episode, we delve into Mark's expertise and explore the best practices for integrating automation in customer support to enhance the overall customer experience. Join us as we uncover the dynamics of this third wave with one of its most seasoned pioneers.

Embracing the Third Wave of Customer Care and the Impact of Generative AI and Conversational Technologies

Mark identifies the current era as the "third wave" of customer care, driven by advancements in generative AI and enhanced conversational capabilities. This wave marks a significant shift toward a “tech first” engagement approach, revolutionizing how companies interact with their customers. As technology continues to evolve, it brings both opportunities and challenges, requiring a strategic balance between automation and human touch.

Despite these technological advancements, one fundamental aspect remains constant: customer expectations. Customers still prioritize having their issues resolved quickly and efficiently while being treated with respect. This unchanging need underscores the importance of designing customer care solutions that meet these timeless expectations, regardless of the technological tools employed.

Leveraging Technology for Better Customer Engagement

The introduction of generative AI and advanced workflows enables two primary approaches: bots and agent assistance. Bots aim to handle customer engagements entirely through automation, providing swift and consistent responses. On the other hand, agent assistance focuses on reducing the cognitive load on human agents, improving their efficiency and compliance during customer interactions. This hybrid approach augments operational efficiency and ensures complex issues are seamlessly transitioned to human agents when necessary.

Rethinking customer engagement with a technology-first approach involves meticulous planning and workflow design. Key considerations include determining when to involve human agents and which technologies to implement. Companies must strike a balance between automation and human intervention, minimizing customer friction while maximizing the benefits of both bots and human agents. Measuring customer satisfaction across these different touchpoints is crucial to refining and optimizing the customer experience.

While technology will play a more direct role in customer interactions, human agents will continue to be an integral part of the equation. Augmented by AI, agents will be better equipped to handle complex issues, offering a higher quality of service. The integration of AI in every customer contact is imminent, making it essential for companies to plan and implement these technologies thoughtfully.

Characterized by the integration of generative AI and advanced conversational AI technologies , the third wave of customer care presents exciting opportunities for enhancing customer engagement. By thoughtfully implementing these technologies and maintaining a focus on core customer expectations, companies can navigate this transformation effectively.

Aligning Metrics and Strengthening Customer Satisfaction With Technology-First Approaches

One of the key themes Mark emphasizes is the importance of maintaining consistency in metrics between technology solutions and agent-assisted support. While the terminology may differ, the core objectives remain the same: ensuring a seamless workflow and unified measurement system across both types of interactions.

Issue resolution from the customer's perspective is paramount. A common mistake some companies make is relying too heavily on customer behavior metrics as their primary customer satisfaction (CSAT) measure, which can leave blind spots. Instead, a straightforward approach should be maintained: whether the care is provided through a bot or a live agent, the primary goal is to resolve the customer's issue quickly and efficiently.

Key Metrics for Agent Assist Tools

When implementing AI-driven support systems, several metrics are crucial for assessing their impact on agent performance and overall operational efficiency:

· Speed to Proficiency: This metric measures the speed at which new agents reach a competent level of performance. By tracking how quickly agents become proficient, organizations can determine the effectiveness of their training programs and the support provided by AI tools.

· Average Handle Time (AHT): AHT is a standard metric in customer service, representing the average duration an agent spends on a single interaction. AI-driven assist tools aim to reduce AHT by providing agents with timely information and suggested responses, thus enabling quicker resolution of customer queries.

· Dispersion of Handle Time: Another essential aspect to monitor is the dispersion of handle time, which refers to the variation in handle times among agents. Reducing this dispersion indicates that the performance gap between the best and worst-performing agents is narrowing, suggesting that AI tools are helping to standardize and elevate agent performance across the board.

With a technology-first approach, additional metrics need to be incorporated to reflect the nuances of automated interactions. These metrics, while using new terms, mirror traditional agent-assisted measures:

· Containment Rates: Equivalent to first contact resolution (FCR), indicating the percentage of engagements fully handled by bots without requiring agent intervention.

· Fallout Rates: Similar to call abandon rates, reflecting instances where customers drop out of the workflow.

· Time in the Bot: Corresponds to the level-of-effort metric, measuring the duration and ease of customer interactions with the bot.

While concurrency is less critical in pure automation scenarios, it also remains an essential metric for agent-assisted interactions. Tools such as agent scripting technology and nudging cues play a significant role in helping agents manage multiple engagements simultaneously, thereby enhancing efficiency and customer experience.

By closely monitoring these metrics, organizations can gauge the effectiveness of these tools and make data-driven decisions to further optimize their customer service strategies.

Balancing Voice and Text in Multi-Channel Customer Engagement

A significant aspect of modern customer care is the shift toward multi-channel and omnichannel engagement. Customer interactions now span multiple channels, such as chat, voice calls, SMS, or email. Effective customer service strategies must seamlessly integrate these channels to ensure a cohesive experience.

Customer Preference and Flexibility

A key aspect of successful customer engagement is allowing customers to choose their preferred communication channel. While companies can guide customers toward certain channels, it is essential to meet customers where they are most comfortable. Flexibility in engagement methods fosters a positive customer experience and reinforces the customer's sense of control and satisfaction.

Integrating Voice and Text in Automation Strategies

Despite the rise of digital communication channels, voice remains the predominant medium for customer contact. However, implementing voice automation comes with unique challenges compared to text-based solutions. Voice interactions can be complicated by factors such as dialect variations, background noise, and line quality issues. These challenges require sophisticated technology and careful planning to ensure that voice delivers a reliable and effective customer experience.

This is why it is crucial to include both voice and text options. For instance, a customer might start with a phone call and receive follow-up communications via SMS or email. Ensuring that both voice and text interactions are effectively managed and integrated into the overall customer service strategy is essential for a holistic approach to customer engagement.

Navigating the Challenges of Integrating Technology in Customer Service Operations

Mark identifies a significant challenge in the current wave of customer care: the need to effectively integrate advanced technologies with traditional customer service operations. Operations leaders are experts in managing customer experiences and executing call center operations. They excel in engaging with agents to deliver outstanding customer service. The rapid advancement of technology, exemplified by innovations like ChatGPT, has introduced new complexities that require a strong collaboration with the technology side.

Understanding and Addressing Technology Deficits

Operations leaders know what they want to achieve and have a clear vision of how to implement these changes within their business. In-house technology teams are typically overwhelmed with existing projects and demands, ranging from outdated code and security requirements to compliance with federal regulations. These constraints create a tech deficit, where the timelines for implementing new technologies are tight and often challenging to meet.

Collaboration for Successful Integration

Successful integration of new technologies into customer service operations necessitates close collaboration between operations and technology teams. This collaboration involves:

· Clear Communication: Ensuring that both sides understand the goals and requirements of new technologies.

· Prioritization: Aligning technology projects with business priorities to manage the tech deficit effectively.

· Resource Management: Allocating the necessary resources to address both existing tech demands and the implementation of new solutions.

· Continuous Learning: Encouraging ongoing education and adaptation to keep pace with technological advancements.

Fostering strong collaboration and addressing the tech deficit allows companies to effectively integrate advanced solutions into their customer care operations. This approach ensures that technological advancements enhance the delivery of exceptional customer experiences.

A Three-Step Approach to Implementing Automation in CX

Step 1: Analytics

The first step in implementing automation in customer service is a thorough analysis of current operations. This involves understanding what has traditionally been done by human agents and identifying the major contact drivers.

Mark emphasizes the need to dissect and analyze high-level workflows that agents typically handle. The analytics phase requires breaking down these workflows into detailed steps to identify various elements that need to be automated. This foundational step sets the stage for effective automation by ensuring that all critical aspects of customer interactions are accounted for.

Step 2: Development and Testing

Once the analytics phase is complete, the next step involves developing and testing the automation solutions. This phase is critical as it transforms the insights gained from analytics into actionable plans and functional systems.

In the development stage, teams create a detailed roadmap and development plan. This includes designing the automation workflows and programming the necessary components. Thorough testing follows, ensuring that the automated systems function correctly and meet the desired objectives before they are publicly deployed.

Step 3: Post-Deployment Measurement and Efficiency

The final step in the automation implementation process is post-deployment measurement and evaluation. This phase focuses on assessing the effectiveness and efficiency of the deployed automation solutions.

Once the workflows and automation are live, measure their performance against predefined metrics. This includes evaluating customer fallout rates, understanding why customers abandon bots, and deciding whether to enhance the existing workflows or move on to new contact drivers. The goal is to ensure that the automation achieves its intended outcomes and continuously improves over time.

A key consideration in this process is maintaining consistent metrics across both automated and agent-assisted interactions. Using the same customer satisfaction and success measurements for both automation and human agents helps gain true insights into the effectiveness of the customer service operations.

Practical Examples of Leveraging Automation for Cost-Effective Issue Resolution

Utilizing Interaction Analytics for Quality Assurance and Customer Insights

Mark shares a practical example of how customer care leaders can effectively deploy automation to drive cost-effective issue resolution, emphasizing the importance of interaction analytics. In one of his larger operations, Mark leveraged this for both quality assurance (QA) and voice of the customer (VoC) insights. With a large offshore population, one of the significant challenges was language skills and understandability. By implementing interaction analytics, the company could evaluate tens of thousands of contacts daily and identify representative insights and coaching opportunities in near real-time.

Furthermore, interaction analytics helped in understanding customer reactions to controversial company policies. Instead of relying on filtered reports, agents could directly listen to customer comments, providing unvarnished insights that informed policy adjustments and improved customer relations.

Interaction analytics play a pivotal role in capturing and analyzing customer interactions. These analytics help in understanding customer sentiments, identifying common issues, and ensuring that both self-service support solutions and live agent interactions meet the desired service standards.

Using Metadata to Reduce Churn and Prevent Fraud

Mark provides another example in the context of membership or subscription services, highlighting the proactive capabilities of automation. One of the biggest indicators of churn is customer engagement with the service. By analyzing metadata on service usage, companies can create nudges to encourage engagement before a customer decides to cancel their subscription.

Mark explains that if usage data indicates a customer is not engaging with the service, automated nudges, such as order status updates, can be sent to re-engage them. This proactive approach can significantly reduce churn and improve retention.

In product-based services, automation extends beyond simple order status updates. Mark illustrates how bots manage complex scenarios like late or lost shipments. A bot informs customers about their order status and, if an issue arises, handles it seamlessly.

For example, if a product is delayed, especially with expedited shipping, the bot offers concessions, refunds, or replacements based on predefined policies. This comprehensive workflow ensures customer satisfaction even in problematic situations.

Metadata is also used to identify potential risks, such as unexpected locations for customer calls, by analyzing call data and comparing it to expected patterns. In financial services, for instance, automation leverages metadata to enhance security and fraud detection. Automation flags these issues for further investigation, providing a more secure experience for customers.

Steps for Brands to Start or Expand Automation in Customer Support

A well-thought-out approach can save significant time and effort in the long run. Critical considerations during the planning phase include:

1. Transparency: Decide whether to inform customers that they are interacting with a bot.

2. Engagement Duration: Determine how long customers should interact with the bot before being transferred to a human agent.

Once the resource needs are identified, brands must decide how to allocate them effectively. Mark suggests three primary options:

1. Reallocate Existing Resources: Shift staff from lower-priority projects to focus on automation.

2. Request Incremental Resources: Advocate for additional resources to support the automation efforts.

3. Utilize Third-Party Solutions: Engage BPOs, IT, or consulting organizations to advance the automation roadmap.

Mark shares a real-world example from his experience with a large seasonal business, which required doubling the staff during peak times. The challenge was the lack of a training environment, which was crucial for preparing seasonal staff. The company built this environment using the six most common transactions. This approach allowed the seasonal staff to practice without the anxiety of live customer interactions, resulting in a 30% improvement in velocity to proficiency.

Starting or expanding the use of automation in customer support involves strategic planning, resource allocation, and leveraging third-party solutions when necessary. Taking a planful approach and addressing key considerations upfront allows brands to integrate automation effectively, leading to improved customer experiences and operational efficiencies.

What Mark Does for Fun

Mark is a dedicated college football and Major League Baseball fan. In the fall, you can find him tailgating at Husky Stadium in Seattle with his family. He also loves traveling in his RV, combining his passion for sports with the freedom of the open road. Mark also works out to stay energized and focused.

To learn more about Mark, connect with him on LinkedIn.

Watch the video here.

Read the blog post here.