Introduction
We have been on the cutting edge of technological innovation for the last two decades. In the early 2000s, RPA emerged, and AI went mainstream by the early 2010s. With the launch of ChatGPT in November 2022, generative AI (Gen AI) has become indispensable in our daily lives. As technology has evolved, its impact has expanded from automating blue-collar tasks to transforming white-collar roles, including creative functions like strategy and design thinking—far beyond the back-office automation of the past.
As we adapt to this new landscape, addressing an often-overlooked aspect—change management—is crucial. This tends to be overshadowed by broader goals like ROI. Historically, change management processes have seen a success rate of just 33%, meaning only a third of employees successfully adapted to new roles. In contrast, another third struggled, and the remaining third were deemed unfit for the new environment.
In the era of Gen AI, we must rethink our approach to change management. How can we ensure the success of these initiatives?
Enterprise Gen AI Objectives
When integrating Gen AI into an enterprise, the objectives can broadly fall into revenue impact and productivity/efficiency enhancement. The change management strategy will differ based on the specific outcomes the organization aims to achieve.
Figure 1: Gen AI projects under consideration fall into two buckets based on business objectives
Revenue-impacting Projects
These Gen AI projects are typically spearheaded by CXO-level leadership for revenue-driven initiatives, aligning with the organization’s overarching vision and goals. These projects often focus on high-level outcomes, such as enhancing customer engagement, optimizing sales processes, or improving supplier negotiations. In these cases, the change management strategy often prioritizes the end goal, with less immediate focus on employee impact until later stages.
For instance, a company might implement an NLP-based chat interface to drive customer sales by improving response times and query resolution. Similarly, intuitive Gen AI-powered dashboards could be deployed for suppliers and partners to better assess sales cycles, manage account receivables, and accurately forecast delivery schedules and revenue cycles. A practical example is Walmart’s use of a Gen AI chatbot to negotiate costs and purchase terms with suppliers. In such scenarios, suppliers must be informed that they are interacting with a bot, not a human, as part of the change management process.
In revenue-impacting projects, user interfaces may be revamped to leverage Gen AI, enabling faster decision-making and more efficient Q&A. Change management introduces new ways for consumers to intuitively manage and interact with the company applications. For example, Swiggy, an Indian online food ordering and delivery company, uses Gen AI-generated food images on its platform. A change management strategy was applied to notify users of the transparency of using Gen AI to help them make well-informed decisions.
Productivity and Efficiency Projects
The change management strategy must be more comprehensive and inclusive for projects aimed at productivity and operational efficiency. Gen AI deployment across the company requires careful consideration of how it will affect various business functions and roles while cultivating a continuous learning and adaptation culture. Below are some of the ways to achieve this.
-
- Foundational training: It is essential to provide all employees with basic training to help them understand the broad impact of AI technology and how it can be integrated into their daily tasks. Many companies have rolled out AI tools such as Copilots organization-wide, but they struggle to get employees to fully utilize these tools despite increased IT investments. Foundational training is crucial to ensure everyone can effectively leverage AI.
- Persona-based training: Targeted training is essential for employees directly engaged with AI, such as data scientists, ML engineers, and those utilizing Gen AI tools, like customer service representatives handling complex inquiries or pharmaceutical scientists researching drug compounds. This type of training should encompass certifications and practitioner-level courses, reinforced by structured programs requiring minimum training hours and regular assessments to ensure ongoing competency and proficiency.
- Creating AI champions: While a central AI governance authority typically prioritizes Gen AI use cases, often involving business leaders, technical teams, and risk management, implementation is usually decentralized. In many cases, department heads are responsible for funding their AI initiatives. To foster a culture of AI excellence, it is important to highlight successful teams or groups as AI champions. This showcases best practices and encourages other departments to learn from their peers and confidently embark on their AI journeys.
- Hackathons: Although the final approval for Gen AI use cases typically rests with the AI governance board, it is vital to tap into the creativity and ideas from all levels of the organization. Hosting hackathons can spark innovative ideas for AI applications that might not emerge from top-down initiatives alone. Additionally, involving employees helps align them with the organization’s AI vision, reducing resistance and increasing enthusiasm for adopting the technology.
Case in Point: Merck, a multinational pharmaceutical company, has implemented comprehensive Gen AI awareness programs, resulting in 90% of employees receiving Gen AI training through various initiatives, such as ongoing workshops and boot camps. Merck takes a collaborative approach for technical staff, such as data scientists with foundational AI experience, pairing them with experts in Gen AI on real-world projects. This hands-on experience deepens their understanding and application of Gen AI. For business users, Merck has established targeted Gen AI awareness programs. The data analytics team supplements the awareness programs by working closely with these users to identify and explore potential use cases, providing training on effectively applying Gen AI to their specific business challenges. This approach ensures that technical and non-technical employees can leverage Gen AI in meaningful and impactful ways.
Best Practices
As organizations introduce Gen AI, moving beyond good intentions and implementing a robust change management strategy is crucial to ensure a smooth transition. Business leaders must consider how best to deploy Gen AI in their environment, and adopting the following best practices in change management can significantly aid in this preparation.
-
- Communicate: Traditional change management often emphasizes grand announcements and events highlighting the “go-live” moment, aligning the future digital state with long-term goals. However, with the rapid pace of technological change and continuous organizational transformation, this approach may no longer resonate with employees. Organizations need to reframe their communication strategies to manage the ongoing evolution brought by Gen AI initiatives. Messages should be less aspirational and focused on immediate, tangible outcomes, connecting the user, the technology, and the business.
- Educate: Clearly understand how specific Gen AI technologies will be used and how they will change the way people work. Before fully deploying AI, allow a buffer period for users to familiarize themselves with the technology. Offer multiple opportunities for hands-on experience. Integrate Gen AI into daily tasks and actively communicate these opportunities. Organizations can effectively scale AI literacy by demystifying AI and showing how it can support their roles.
- Upskill/Reskill: Education through empowerment opens doors to new possibilities, but organizations must also build resilience in their workforce to navigate these changes. Prioritize learning programs that develop critical skills for the future as technology continues to evolve.
- Monitor: Establish a feedback loop to maintain an ongoing dialogue with users. This could take the form of surveys or working groups where users can share what’s working and what’s not and suggest improvements. This continuous feedback is vital during the education phase and beyond.
- Refine: Feedback loops are only effective if organizations are willing to listen and learn. Traditional change management often reacts to unexpected feedback by reinforcing the same messages from higher levels. Instead, organizations should pause and consider revising their approach. This might involve revisiting or reprioritizing use cases, allowing more time for education, or providing additional training to support upskilling or reskilling.
- Govern: Gen AI is rapidly evolving, and what works today may not be relevant in a year. Establish a timeline for revisiting your approach. The technology you implement now, along with your change management strategy, must be adaptable to withstand the test of time. Regularly review and adjust your strategy to align with technological advancements and organizational needs.
Conclusion
As AI and generative AI continue to reshape industries in the coming years, organizational leaders must make time today to enable their employees, suppliers, partners, and consumers to embrace future changes. By embracing generative AI strategically, creating and communicating an inclusive view of the value of human expertise in the future, and supporting skill development for today, businesses can navigate change management effectively, seize opportunities, and stay ahead in an era of transformative technology.
By Tarun Mehta, Principal, Avasant and Chandrika Dutt, Associate Research Director, Avasant