Thursday, December 26, 2024
This article summarizes insights from the From promising to productive: Real results from gen AI in services McKinsey report. To receive more report summaries powered by Finnt, complete the form below.
InsightDetailsAI at scale11% of companies worldwide are using generative AI at scale.GenAI scaled in operations3% of large North American and European companies have scaled a generative AI use case in operations.GenAI solutions in finance 202311% of organizations were piloting generative AI solutions in finance functions.GenAI solutions in finance 202445% of organizations are now piloting generative AI solutions in finance functions, with 6% having achieved scale.
Generative AI (GenAI) represents a transformative opportunity for companies across various sectors, offering the potential to significantly enhance operational efficiency, drive revenue growth, and improve customer experiences. A few pioneering companies have already begun to capture substantial value from GenAI, attributing more than 10 percent of their Earnings Before Interest and Taxes (EBIT) to its use. This success underscores GenAI's potential to not only optimize existing processes but also to create new value streams and competitive advantages.
The companies capturing real value from GenAI span multiple industries, including finance, telecommunications, and consumer goods. These organizations are characterized by their early adoption of GenAI technologies, strategic implementation across operations, and commitment to innovation. They have successfully scaled GenAI solutions in specific domains such as customer service, risk assessment, and financial planning, demonstrating significant improvements in efficiency, productivity, and customer satisfaction.
To replicate the success of these early adopters, companies should consider the following insights:
The necessity of building a strategic approach to GenAI implementation cannot be overstated. By starting with lower-risk use cases, companies can quickly learn and iterate on their GenAI strategies without significant upfront investment or risk. This approach allows organizations to build a solid foundation of knowledge and experience, which is crucial for tackling more complex, higher-risk use cases in the future. Moreover, a strategic approach ensures that GenAI initiatives are closely aligned with business goals, thereby maximizing the return on investment and ensuring sustainable, long-term success.
The global bank's strategic deployment of generative AI (GenAI) serves as a compelling example of how financial institutions can leverage this technology to enhance operations and customer service. This case study outlines the bank's approach from identifying potential domains for GenAI implementation to realizing tangible benefits in customer experience and operational efficiency.
Initially, the bank considered 23 potential domains for GenAI application, spanning across various aspects of its operations. These domains included customer service enhancements, risk assessment optimizations, and process automation in both front-office and back-office functions.
After a thorough assessment of business impact and technical feasibility, the bank narrowed down its focus to two primary domains:
In both selected domains, the bank identified key commonalities in GenAI application that guided its strategic approach:
The implementation of a GenAI-powered chatbot in the consumer banking contact center brought several advantages:
The chatbot implementation led to notable improvements in customer experience by:
Building on the success of the chatbot in the consumer banking domain, the bank applied the lessons learned to its KYC processes within corporate and investment banking. The GenAI technology underpinning the chatbot was adapted to create a "smart, virtual agent" that automated significant portions of the KYC process. This application demonstrated the bank's strategic approach to GenAI deployment, focusing on scalability and cross-domain applicability. The virtual agent facilitated a more efficient data collection and verification process, significantly reducing the manual effort required from both customers and bank employees. This not only improved the efficiency of the KYC process but also enhanced the overall customer and employee experience by streamlining previously cumbersome procedures.
This case study explores how a leading North American telecommunications provider reevaluated its customer journey with generative AI (GenAI) input, leading to significant operational improvements and enhanced customer satisfaction.
The telecommunications provider embarked on a comprehensive reevaluation of its customer journey, focusing on identifying and addressing invisible pain points. By leveraging GenAI technologies, the company was able to gain deeper insights into customer behaviors and preferences, enabling a more customer-centric approach to service design and delivery.
A specific area of focus was the customer journey for changing a mobile phone number. The process was mapped out in its entirety, revealing that it was overly complex and a significant source of frustration for customers. GenAI tools were employed to analyze the journey, identify bottlenecks, and redesign the process to be more intuitive and user-friendly.
As a result of these improvements, the company experienced a substantial reduction in total call volume to its customer service centers. By addressing the root causes of customer inquiries and streamlining the mobile number change process, many issues were resolved without the need for direct customer service intervention, leading to a roughly 30% decrease in call volume.
The implementation of GenAI solutions also led to a significant decrease in average handle time for customer service calls. With more efficient processes in place and GenAI-powered tools to assist customer service representatives, the time required to resolve each call was reduced by more than one-quarter.
The strategic use of GenAI in reevaluating and improving the customer journey led to notable improvements in customer care. Customers benefited from faster, more efficient service, reduced wait times, and a more seamless experience when interacting with the company.
Another key outcome of the GenAI-driven process improvements was a rise in first call resolution rates. By better understanding and addressing customer needs on the first contact, the company was able to improve resolution rates by 10-20 percentage points, enhancing overall customer satisfaction.
The reevaluation of the customer journey and the integration of GenAI tools also had a positive impact on agent capacity. By streamlining processes and reducing the complexity of technology systems, agents were able to handle inquiries more efficiently. This optimization, combined with a reduction in total call volume and average handle time, allowed the company to reallocate resources more effectively, improving the overall productivity of the customer service function.
Scaling generative AI (GenAI) across an organization involves more than just technological implementation. It requires a holistic approach that encompasses governance, performance, change management, and a culture of continuous innovation. Here are the four secrets to successfully scaling GenAI:
Effective governance is crucial for the successful deployment and scaling of GenAI. This includes establishing a cross-functional steering group, appointing a dedicated AI governance officer, and developing clear AI guidelines and policies. A structured decision-making process for evaluating GenAI proposals ensures that initiatives are aligned with the organization's strategic objectives and comply with regulatory requirements.
Scaling GenAI requires modernizing the performance infrastructure to accommodate the new operational strategy. This involves redefining metrics to reflect GenAI's impact, implementing a disciplined review process, and leveraging data analytics to tailor coaching and training programs. A focus on performance ensures that GenAI initiatives contribute positively to the organization's overall productivity and efficiency.
Change management is a critical component of scaling GenAI, as it addresses the human element of transformation. Effective change management strategies include clear communication about the benefits and implications of GenAI, upskilling and reskilling initiatives to prepare the workforce, and fostering a culture that encourages innovation and adaptation. Engaging employees in the process helps to mitigate resistance and build a supportive environment for GenAI adoption.
A culture of continuous innovation is essential for keeping pace with the rapid advancements in GenAI technology. Celebrating successes, sharing best practices, and maintaining a "buy and build" mindset are key to fostering an environment where innovation thrives. Encouraging frontline workers to contribute ideas and reexamine assumptions about the role of partners and vendors in sourcing innovation ensures that the organization remains at the forefront of GenAI development.
By focusing on these four areas, organizations can effectively scale GenAI initiatives, driving significant improvements in efficiency, customer experience, and competitive advantage.
The evolution of generative AI (GenAI) technologies, including large language models (LLMs) and retrieval-augmented generation (RAG) systems, has significantly impacted various industries by automating complex tasks and processes. However, these technologies have limitations that can hinder their effectiveness in certain applications. Conversely, multiagent systems offer a promising alternative, overcoming some of these limitations with their unique advantages.
Complex Processes Struggle: LLMs and RAG systems often find it challenging to handle complex processes that require multiple steps or decision-making points. Automating only parts of such processes can result in limited benefits, as the entire workflow might not be sufficiently streamlined.
Proneness to Error: Without the ability to cross-verify information or understand context deeply, LLMs and RAG can generate outputs that are inaccurate or not fully aligned with user intentions, leading to potential errors in critical applications.
Limited Beyond Text: LLMs are primarily designed for text-based applications, limiting their utility in scenarios that require understanding or generating non-textual data. While RAG systems can use multiple data sources, they are expensive to build and scale, especially for applications requiring complex data interpretation.
Quantitative Analysis Limitations: LLMs have limited capabilities in performing quantitative analysis, making them less effective in applications that require numerical reasoning or financial modeling.
Complex Workflow Execution: Multiagent systems excel in executing complex workflows by breaking down processes into smaller, manageable tasks. This approach allows for more reliable task execution and the ability to handle intricate processes that LLMs and RAG struggle with.
Self-Correction and Improvement: By incorporating feedback loops, multiagent systems can learn from their actions, correct errors, and improve over time, leading to higher quality outcomes and reduced error rates.
Example: A North American bank implemented a multiagent system to automate the writing of credit risk memos. This system was able to identify the correct data sources, ingest up-to-date data, and integrate qualitative and quantitative insights. As a result, credit decisions became 30% faster, relationship manager productivity more than doubled, and revenue per relationship manager increased by 20%.
Integration of Qualitative and Quantitative Insights: Unlike LLMs, multiagent systems can effectively integrate both qualitative and quantitative data, providing a more holistic view of the problem at hand. This capability is particularly valuable in fields like finance, where decisions often rely on a mix of numerical data and qualitative assessments.
Cost-Effective Scaling: Multiagent systems can be more cost-effective to scale compared to RAG systems that require extensive data sources and processing power. By focusing on specific tasks within a process, multiagent systems can be deployed more efficiently, reducing the overall cost of implementation.
In summary, while LLMs and RAG systems offer significant benefits for text-based applications and data retrieval tasks, multiagent systems present a more versatile and effective solution for complex process automation, self-improvement, and integrating diverse data types. These advantages make multiagent systems a cutting-edge choice for organizations looking to harness the full potential of GenAI.
The exploration of generative AI (GenAI) across various case studies and theoretical frameworks has underscored its transformative potential in enhancing operational efficiency, customer experience, and strategic decision-making within organizations. Key insights from the analysis include:
In conclusion, the strategic deployment of GenAI presents a significant opportunity for organizations to revolutionize their operations, enhance customer interactions, and achieve substantial business growth. By adopting a disciplined, strategic approach to GenAI implementation and fostering an environment conducive to innovation, companies can unlock new value streams and establish a strong competitive edge in their respective industries.
Don't waste your finance teams on repetitive and manual tasks. Let AI do the work.