[[Audio 1]] [[DownloadsSidebar]]

There should not be any doubt at this point that customer experience is a tremendous factor of growth. When we analyzed the financial performance of 75 consumer-facing companies between 2016 and 2021, we found that those with the highest customer experience ratings achieved twice the revenue growth of their peers.[[footnote 1]]

[[MostPopularArticles 5]]

This isn’t magic. Companies that look after their customers tend to acquire more of them, sell more to each one, and keep those customers for longer than companies that don’t.

Nor is it easy. The factors that propel great customer experience—such as responsive, personalized service—have traditionally been expensive to offer. They depend on large numbers of skilled and motivated frontline employees, supported by managers and specialized experts, along with extensive technical resources continually analyzing insights related to the customer base.

Right now, few organizations are in a position to make large investments in service function expansion. After two years of high inflation and rising interest rates, many companies are looking to cut costs across their operations, instead of doubling down on these capabilities. Executives have lower optimism for consumer demand, which makes them cautious about new capacity investments. And even businesses that want to scale up their service teams are finding it difficult to hire and retain additional staff.[[footnote 2]]

There are technologies that can help

In this environment, many companies are looking to technology to solve the cost and customer experience paradox. Recent developments in digitally enabled customer service promise dramatic improvements in efficiency and productivity, while simultaneously enabling the kind of deep personalization that can turn an everyday service interaction into a great experience. Generative AI (gen AI), for example, is helping companies craft targeted communications based on individual customer preferences and purchase history. Interactive gen-AI-powered chatbots or interactive-voice-recognition (IVR) systems can handle complex queries in a fast and user-friendly way. As IVR, chatbot, and virtual-assistant technologies advance, customer experience may eventually evolve to a channel-agnostic, AI-based support. Behind the scenes, digital-workforce-management systems can help companies adjust service team schedules to meet fluctuating demand, while service agents using AI tools can provide customers with faster, more accurate answers.

As they race to implement these new technologies, companies should avoid the pitfalls that have hampered previous digital-transformation efforts. In a 2022 McKinsey survey of senior business leaders, nine out of ten said that their organizations had pursued at least one large-scale digital transformation in the previous two years. Overwhelmingly, those transformation efforts underperformed, generating only 31 percent of expected revenue increases and only 25 percent of expected cost savings.[[footnote 3]]

When companies reflect on the issues that held their technology transformations back, a few common themes emerge. Some are common to any business change effort, such as insufficient commitment or alignment from senior leadership—which can slow down decision making and make it difficult for the transformation to build and sustain momentum. Another is the tendency to chase shiny objects but fail to build the foundational capabilities needed to do so well. Many companies’ transformations—whether they're for microsegmentation, front-to-back design, talent, or the operating model—fail because of these issues.

Some themes are unique to digital-transformation efforts. One is an overabundance of caution. Some companies take a piecemeal approach, introducing new technologies incrementally across their operations, with each change designed to address a specific issue or pain point. Working this way helps companies manage costs and risks, but it can mean overlooking opportunities that require large-scale change. And those may be the ones that add the most value in the long run.

Other companies do the opposite and see a specific technology as the silver bullet for all their operational challenges. They try to fix everything with a single tool. This approach is doubly ineffective. Not only does it fail to fix the underlying processes, but it also forces teams to adopt tools that may not effectively address the challenges they are facing.

Building an AI-enabled services organization

In the coming wave of AI-enabled digital transformations in services, companies can avoid these pitfalls by getting three essential things right. First, they can deploy the right technologies at the right time and in the right way. Second, they can integrate the use of technology and the application of proven business improvement levers. Finally, they can equip their people with the skills and capabilities they need to embrace technology.

Right technologies, right time, right way

Effective, digitized services should not be built on a single technology, but rather assembled from a toolbox of technological components. As with any toolbox, users need to choose the right tool for each task. Gen AI technologies have given companies a powerful new way to summarize and process messages and other natural language inputs, for example. Where a task requires the processing of structured data, however, robotic process automation may be a faster, cheaper, and more reliable solution. Deploying technologies in an integrated way, in fact, can help amplify the impact, as the below examples of organizations illustrate.

[[globalsidebar QB]]

Timing plays an important role in technology adoption. Each phase of a transformation process requires a different focus and is best supported by different technology tools. These phases typically include diagnosis, planning, and implementation. The diagnostic phase focuses on understanding the full potential of the organization. The planning phase is about optimizing operations and developing strategies for moving forward. The implementation phase focuses on scaling, sustaining, and refining operations.

Finally, it’s imperative to deploy selected technologies with an appropriate operating model. This includes decisions such as whether to proceed with a proof of concept or a full build and whether to acquire licenses or use the services of a systems integrator. Introducing technologies in the right way ensures a smooth implementation and minimizes disruptions. During the diagnostic phase, for instance, insights gained from processes may lead to the deployment of hyper-automation technologies like optical character recognition (OCR) and intelligent document processing, or other tools like machine learning, low- or no-code platforms, and digital twins may be required to address operational needs. In addition, selecting a business unit with sufficient opportunities and leadership commitment can help overcome potential challenges encountered along the implementation journey.

One North American wealth management company used a series of technologies to achieve significant recurring efficiency savings. It automated five end-to-end operational processes, starting with document extraction, verification, and data ingestions for all those processes. The company began by conducting a top-down, role-based automation potential assessment and applying targeted task-mining tools to build a view of the manual effort required to process thousands of instruction documents and forms every year. It then developed a proof-of-concept system that used OCR technology to classify and extract handwritten data, combined with robotic process automation to validate that data and insert it into downstream systems. The new system included machine learning technology so that it can continually adjust its own internal processes for improved performance.

The new fully automated system, which was scaled and rolled out across the company, allowed the company to restructure its back-office operations and eliminate downstream errors. The company reimagined its unit for central mail processing and data entry as a digital-data-quality control center and reassigned more than 70 percent of the staff to other tasks. The change also had a direct positive impact on the customer experience as faster, digitized processing reduced end-to-end times for new account applications and captured any errors nearly instantaneously, often while the end customers were still present.

Fix and reimagine processes as they are digitized

Technology can’t fix problems caused by broken service processes, and there’s no point in creating a fast, streamlined, and efficient process that still provides unsatisfactory customer outcomes. That’s why successful tech-enabled transformations require the seamless integration of technology and traditional business improvement levers.

The first step is to understand and simplify current processes. This involves managing demand and eliminating unnecessary tasks using process mining, discovery, and documentation tools. By creating a baseline for the current state of processes and streamlining policies, inputs, exceptions, and services, complexity can be dramatically reduced.

Next, it is critical to orchestrate and digitize processes. By redesigning workflows and creating structured digital outputs, handoffs can be reduced and information can flow more smoothly. The use of tools such as natural language processing (NLP), OCR, chatbots, and web forms helps generate digital, structured inputs.

Once simplified and digitized, automation and integration of digitized processes are the next essential steps. Automation of specific process components can be achieved by designing and configuring bots, APIs, and microservices, especially for activities governed by known rules. Machine learning algorithms can be trained to handle activities with complex or unknown rules. Companies should ensure that the foundational layer can support these technologies.

Finally, reorganizing and empowering the workforce completes the transformation cycle. Some of this change is structural, optimizing the design of the organization and improving efficiency through centralization or offshoring. The rest is about mindsets, skills, and capabilities, which we discuss below.

A multinational professional services firm took an end-to-end approach to improving its back-office technology support center. It began with in-depth, AI-powered analysis of thousands of data points across four channels including call, chat, email, and a case management tool. The firm processed hundreds of hours of agent activity to understand how they spent their time in different applications and identify critical pain points. These insights were used to improve performance across three dimensions: customer experience, process, and people.

The company used a combination of gen AI, conversational AI, and task-mining technologies to find ways for tech support staff to use their time more effectively. It found that top-performing agents would use “dead time” during calls to review chats and documentation relating to other cases, for example. Using task mining along with traditional customer care diagnostics, the company also identified various shortcuts and standard techniques used by the best agents to resolve common problems. Sharing those approaches helped other agents fix customer issues more quickly, reducing average handling time.

Equip your people with the right capabilities and skills

People are the secret ingredient in any digital transformation. McKinsey research on successful digital implementations found that top performers were 1.6 times more likely than others to have achieved their people-related goals,[[footnote 4]] which include large-scale capability-building efforts, improvements to employee experience, and effective talent management.

As technology advances, the importance of talent management and capability building rises. Organizations that have already made progress in the implementation of AI expect to reskill large parts of their workforces over the next three years, with 73 percent aiming to boost the skills of more than 30 percent of their people in that time.[[footnote 5]]

A digitally enabled workforce needs the right tools, together with the skills, workflows, and support resources to use those tools effectively. Most service employees will not have advanced-computer-programming capabilities, for example, but low- and no-code digital platforms can allow nonspecialists to analyze data or automate tasks.

Technology is also changing the “field and forum” methods that companies traditionally use to upskill their staff. Forum-based learning is transitioning from sporadic events to continuous learning experiences. Previously, capability building relied on one-off, event-based workshops, which often lasted a week and were held annually. However, this approach has shifted toward tech-enabled methods, featuring microlearning sessions and immersive online modules. These resources are personalized using extensive libraries and advanced technologies.

Similarly, field-based learning is undergoing a transformation from observation-based practices to data-enabled ones. In the past, on-the-job coaching was conducted through sample observations, such as call-listening sessions. Now, coaching is becoming more precise and dependent on data, using analytical tools powered by NLP-based machine learning algorithms. This allows for tailored coaching opportunities based on individual performance metrics.

An engaged, motivated service workforce is fundamental to long-term performance. The time-tested principles of lean management—such as a relentless focus on eliminating waste and defects, and making improvements based on frontline employee insights—are still essential in the AI-enabled world. The only difference is in the tools available to gather insights and make improvements. Low-tech solutions such as job aids and policy changes can now be complemented by digital solutions built, tested, deployed, and maintained by frontline employees.

Advanced digital technologies could unlock an elusive goal for services organizations, allowing them to offer better customer experiences at lower cost. Organizations are already exploring the use of innovative AI technologies through collaboration between their IT and operations functions. They foresee a world where decisions and customer journeys along the end-to-end value stream are augmented by AI, gen AI tools enable new levels of intelligence and personalization for those journeys, and smart, real-time support technologies transform service agents into “superhumans in the loop” who are focused on knowledge and value creation.

The journey to a tech-enabled service model is fraught with challenges, however. To make it work, companies should select the right technologies, implement them at the right phase of the transformation, and apply them in the right way.

They should think beyond the technology, too, applying digital tools alongside traditional transformation and improvement methodologies and empowering their teams to embrace continuous improvement. If done right, digitally enabled service excellence can create a virtuous circle, freeing up resources to invest in service improvements and innovations that differentiate the organization from its peers, induce growth, and contribute to sustainable competitive advantage.