On Change Management in a Company in the Age of Artificial Intelligence

The chief customer and digital officer at US energy company Ameren believes that successful large-scale transformation requires implementing robust change management methods.

Bhavani Amirthalingam’s career includes more than 25 years of experience in key technology and leadership roles across multiple industries and geographies. She spent 15 years at World Wide Technology, a high-tech, fast-growing company, serving as CIO and vice president of customer solutions and innovation. She then led the digital transformation of Schneider Electric, a Fortune 100 energy management company. She most recently served as executive vice president and chief customer and digital officer at Ameren, where she joined in 2018 as senior vice president and chief digital and information officer, and was promoted to chief customer officer in 2023.

In a recent episode Tech Whisperers podcast Amirthalingam shared her career path, the hallmarks, and the leadership philosophy that helped her land the role of Chief Digital and Information Officer, then the Chief Customer Officer, and an offer to join the board. After the podcast, we chatted a bit more about what it takes to successfully lead a large-scale transformation in an organization, and how leaders and organizations should use AI and data in the future. Below is a transcript of that conversation.

Interviewer: Given your experience helping large organizations transform their customer experience and increase digitalization across the enterprise, what advice do you have for other leaders embarking on large-scale transformations?

Bhavani: For such transformations to be successful, it is critical that cross-functional teams come together and set clear, overarching goals. Sometimes organizations reduce the entire transformation to technology. But this is not a technology initiative or an operational initiative. It is a company or customer initiative.

First, an important element in carrying out such transformations is the selection of two or three people – depending on the organization – who are responsible and accountable for the transformations, both at the local level and at the very top.

The second point is that most organizations underestimate the change management itself, which is necessary for successful transformation. They think of change management as something that happens at the end of the project, and this approach is doomed to failure. In fact, you start change management at the very beginning, focus on it throughout the project and beyond.

Change management starts with identifying the reasons for the change. This is where you need to involve those who are doing the actual work with their hands; you can also involve customers to get their perspective and needs, and then keep them engaged throughout the process. You need to identify and create this “focus group” early on, making sure there are different points of view. There will probably be at least one person who is enthusiastic about the change because they want it to happen. Then you may want to invite “difficult” participants into the conversation as well – this will give you an idea of ​​what concerns they have and why.

Active support for change management implementation should start from day 0, not day 9 or day 10, because that's where most things fall apart. Knowing your area of ​​work is essential – and it's not just the responsibility of the technical team; it's the responsibility of the teams and managers that will be affected by the project.

People can be fascinated by technology. They've been waiting for it and they've been hungry for it, and so you move forward. But you're not thinking about change management and the fact that you're going to change something that people have been doing a certain way for years.

Interviewer: What other blind spots are there that can undermine success?

You can have the right goal, the right initiative, great teams, everyone is aligned, positive changes are happening, and implementation has already begun – and yet you may still fail to realize the value of this transformation.

Value realization is a really important aspect of transformation, so it's important that you define what success looks like up front. What are the key performance indicators that we're going to be looking at to say, for example, we were at X, we needed to achieve Y, and we did that. What does that time frame look like? And what does that path look like?

And when you're talking about global Fortune 500 companies, there are many layers of complexity. You have to understand the organizational dynamics and culture. For example, the organization may be matrixed, so you have to figure out what it takes to get everyone on the same page.

You also need to make sure that the CEO is engaged and that it is not just left to the technology or operations leader. If you want to achieve scale and significantly increase the digitalization rate of the organization over the long term, CEO engagement is a must for success and value realization.

Interviewer: You've done a lot of interesting work in your career using data and AI. What opportunities do leaders have to use AI to impact their employees and improve customer service?

Let me start by saying that artificial intelligence is not new. It emerged in the 1950s with machine learning. Using data and algorithms to imitate human learning emerged in the 1980s, and evolved into deep learning in the 2000s. Acceleration in computing has led to the creation and scaling of large language models, which have now democratized artificial intelligence.

I had the opportunity to use AI to improve customer service by:

• providing customers with more detailed information about their energy consumption;
• improving asset performance through prediction, failure prevention and targeted maintenance to improve reliability;
• improving customer and employee safety by detecting gas and water heater leaks;
• reducing bills for low-income customers by optimizing voltage in circuits and many other similar use cases.

My advice to leaders:

• identify areas with the greatest potential and impact,
• assess data readiness,
• create or implement existing solutions using artificial intelligence,
• and also make sure you rethink how people will work differently with these new capabilities, from the very beginning of your initiative.

There are many ways to approach generative AI. First, you can be a receiver – understanding the built-in capabilities of the software platforms you own and identifying how you can use them to benefit your organization. An example would be the platforms you use to develop software that now have AI capabilities – using them to optimize the development and capabilities of your collaboration suite, CRM, HR applications – the list goes on.

The second approach might be the “modeler” approach: where you create a private and secure instance of an existing large language model using one of the hyperscalers – essentially your own version of ChatGPT for your company, fine-tuned to your data to enable specific scenarios for your business. You can create “co-pilots” for different functions in the organization. A third option might be to create your own large language model, which can be expensive and complex, but which can create distinctive value for your business.
Two of the most common use cases for AI across industries are improving developer productivity and customer service representatives in call centers. An example would be debriefing after a call. You’re increasing productivity so you can spend time talking to the customer and doing more important things that only you can do. The key is to identify the use cases that have the most value and impact on the business and understand how you’ll work differently to take advantage of those opportunities. Cybersecurity principles should be a guideline when creating those opportunities.

The human factor will matter more. It is very important to remember this and to improve the skills of our employees. Artificial intelligence will not put everyone out of work, but as Karim Lakhani from Harvard Business School said, “AI will not replace humans, but humans with AI will replace humans without AI.”

Tell us about how leaders should think about the role of data quality when implementing AI.

Data quality is the cornerstone of effective AI implementation. Without it, scaling AI solutions is like building a house without a solid foundation. Leaders must prioritize investments in data quality and governance. Training algorithms on quality data takes time, but it is necessary to achieve desired results at scale. Pilots often demonstrate the feasibility of scaling, but not all initiatives will make business sense. It is critical to design a sustainable architecture with the end goal in mind and ensure that the scaling process is aligned with business goals.
Leaders should view data quality as a strategic asset. High-quality data enables efficient training of algorithms, which leads to more accurate and reliable AI applications. It is essential to establish a robust data governance system that ensures data integrity, security, and compliance. Such a foundation supports AI systems that can adapt and scale as business needs evolve.

How do you think artificial intelligence will impact different aspects of business in the future?

I think every aspect of business will be affected in some form. As an analogy, think about how the internet, mobile, and cloud technologies have changed the way we work and live over the last 25 years. Artificial intelligence will have an even greater impact on both our daily lives and business; and I think the changes will happen even faster.

I’m really excited about the possibilities of solving big, complex problems that can improve human life in our lifetimes, whether that’s solving climate change or finding a cure for cancer. Think about the pharmaceutical industry and how long it takes to research and develop drugs. I think that using AI will reduce that time frame dramatically. In terms of product innovation, the service or product you’re offering may be significantly changed or even completely disrupted, depending on the industry. So it’s really important that you also look at the “product” you’re delivering and think about it from that perspective.

It's really interesting to hear you talk about AI with such energy and optimism, because many people are much more wary of it.

I am truly excited about the potential of AI to tackle diseases that we currently cannot. However, this progress does not come without risks. Cybersecurity and data privacy are critical areas that need to be negotiated. The energy consumption of data centers that power AI challenges our goal of achieving net zero energy. Despite these challenges, I am confident that we will develop the necessary safeguards and continue to use AI for the benefit of humanity. I believe that AI will help solve some of the problems it has created.

Why does a company often encounter resistance when trying to implement global changes? Is resistance to change good for a company and how can it be dealt with when implementing a strategy? Let's talk about it at an open lesson on July 11.