What McKinsey learned while creating its generative AI platform
Jan 22, 2025
auto_awesome
Erik Roth, a senior partner at McKinsey and global leader in growth and innovation, discusses the creation of their generative AI platform, Lilli, named after a pioneering female innovator. He delves into the balance of rapid tech development and user experience. Roth emphasizes the shift in consulting towards actionable insights, the need for high-quality data, and evolving hiring practices. He also highlights the importance of AI in education for business leaders and the challenges of integrating technology safely while avoiding legacy pitfalls.
McKinsey's generative AI platform, Lilli, emphasizes user-centric development, balancing rapid technological advancements with essential user feedback throughout the process.
The evolution of McKinsey consultants towards technology enablement suggests a transformative shift in skill requirements, focusing on delivering insights and promoting diversity in practices.
Deep dives
Challenges of Building Generative AI Solutions
Building a generative AI solution involves balancing the pace of development with the need for user feedback. Rapid building through coding and technology assembly often outstrips the learning curve that comes with waiting for users to engage with and provide insights on the product. The process starts with establishing a clear understanding of user needs, which can be informed by observational research and maintaining a user-centric approach throughout development. This iterative process allows for continuous improvement and adjustment based on real-world usage and user experience.
Distinctive Features of Lilly
Lilly, as McKinsey's generative AI solution, stands out from other platforms through its tailored approach to client service and knowledge management. It employs a combination of various technologies, creating an orchestration layer designed specifically for the organization’s needs, which ensures compliance with regulations and security protocols. Unlike traditional retrieval-augmented generation systems, Lilly personalizes responses based on McKinsey's unique methodologies and client contexts, enhancing the relevance and usability of the information provided. This targeted development process reflects McKinsey's commitment to delivering distinctive client service while supporting their internal knowledge-sharing goals.
Embracing Technology for Future Consultants
The future of the McKinsey consultant will likely involve a significant increase in technology enablement, allowing for a shift in focus from tedious analytical tasks to activating insights and delivering value to clients. Generative AI tools will streamline data analysis, potentially transforming the skill sets required from new hires and altering the hiring process itself. Additionally, as consultants become more tech-savvy and empathetic in their interactions, there is a hopeful outlook toward diversity and inclusive practices in the workplace. Organizations must also prioritize data architecture to maximize the potential of generative AI, ensuring that data integrity supports successful application and development.
Nearly one hundred years of McKinsey’s insights and knowledge serve as the source material for the firm’s AI platform, “Lilli.” In this episode of the At the Edgepodcast, Erik Roth, a McKinsey senior partner and global leader of growth and innovation, joins senior partner Lareina Yee to discuss how Lilli was developed and thoughtfully executed—and why immediate access to the firm’s intellectual property has transformed the way McKinsey serves its clients.