With operations in over 200 countries and territories and a portfolio of beloved brands like Lay’s, Gatorade, Quaker, and Pepsi, PepsiCo is more than just a household name — it’s a global powerhouse in food and beverages, generating over $91 billion in net revenue in 2024. As the company accelerates its digital transformation journey, data and AI are taking center stage in enabling innovation, efficiency, and growth across every corner of the enterprise.

In this first installment of a three-part series, Amlan Maitra, SVP and Head of Enterprise Data at PepsiCo, speaks with Clyde Gillard, North American AI GTM Leader at HPE, about how the organization is building a robust enterprise data foundation — harmonizing legacy systems, applying a “silver layer” architecture, and creating reusable, governed data assets that power both traditional machine learning and emerging GenAI use cases.

From demand forecasting and store clustering to digital twins in manufacturing and generative AI (GenAI) copilots for developers, PepsiCo’s data strategy is deeply embedded in its pursuit of top-line growth and operational excellence.

With over 20 years at the company, Maitra and his team have played a pivotal role in shaping PepsiCo’s enterprise data strategy, which spans from “seed to shelf” and supports operations across all geographies and business units.

The conversation reveals not only the complexity of organizing and governing data across a global organization but also the critical role of executive sponsorship, platform modernization, and AI-first thinking in making that strategy a success.

Edited Excerpts

Q: You’ve been with PepsiCo for over two decades and have seen a lot of transformation. Can you tell us about your background and current role at the company?

Our vision is to become the global leader in convenience, snacks, and beverages by winning with purpose.

To that end, as you can imagine, data, analytics, and increasingly AI are important enablers of realizing this vision — helping us become more consumer-centric, grow our scale, and extend our brands into new channels and geographies as we propel our growth.

I’m currently part of our Data, Analytics, and AI team, which sits within our Strategy and Transformation organization. My team and I are responsible for data engineering, data management, and all supporting platforms including global IT data and integration.

We are a centralized organization, so my remit includes building PepsiCo’s enterprise data backbone and enabling AI and analytics solutions across the entire value chain, from seed to shelf, and for all our business and operating units around the world.

Q: You mentioned that everything starts with data and that PepsiCo has built a centralized global strategy for managing it. Given the scale, how do you organize and govern all that data to support global operations, innovation, and compliance?

Everything begins with data. One of the things we recognized early on, as the company embarked on its digital transformation journey, was that data had to be at the core. It was critical to have a robust data foundation and this was supported by the senior-most leadership.

We’ve been on a multi-year journey to build what we call the Enterprise Data Foundation, bringing together internal and external data into a common platform infrastructure. We follow what’s often called a “medallion architecture” in the industry, harmonizing and normalizing data into what we refer to as a common silver layer, which can then be used across multiple digital and analytics use cases.

This approach drives reuse — source, transform, and curate it once, and then use it many times. Increasingly, we’re leveraging this foundation for AI use cases as well.

The program also involves consolidating legacy data warehouses and lakes, modernizing our platform infrastructure, and placing strong emphasis on data governance. We never had a centralized, dedicated focus on governance like this before. It ensures the data we’re building is high-quality, properly cataloged from both technical and business metadata perspectives, easily findable, and has clear lineage. This is truly the backbone of all our digital, analytics, and AI initiatives.

Q: You mentioned the idea of a “silver layer,” a concept that sounds especially useful for companies with legacy systems. Given that PepsiCo wasn’t born in the cloud, how has this approach helped you prepare for AI? And what are some of the key AI use cases you’re focused on now?

From a traditional AI standpoint (think about machine learning and optimization capabilities), we’ve applied these in use cases like demand forecasting, marketing mix modeling, audience generation, and store clustering.

We also use optimization for solutions like customer promotion calendars. These use cases have been around for a while and are powered by the Enterprise Data Foundation. The data governance and quality we’ve focused on are critical to making those advanced machine learning models work well.

Now, with the rise of GenAI, we’re taking an AI-first approach to solving problems. On the tech side, we’re already using GenAI to drive greater efficiency through automation across the software development lifecycle. For example, we’re auto-generating code for data management tasks and automating data quality checks.

We’re looking to embed GenAI across every step of the data and analytics value chain. We’re already seeing benefits and are excited about scaling further.

As for use cases, we’re applying GenAI across the board. In product innovation, it’s helping us accelerate development cycles. In marketing, it’s driving hyper-personalized campaigns through faster consumer insights. In manufacturing and warehousing, we’re experimenting with digital twins to transform operations.

We’re also looking at how GenAI can support enabling functions like HR, finance, legal, and procurement. It’s a comprehensive effort aimed at creating efficiency, but also unlocking top-line growth opportunities.

CDO Magazine appreciates Amlan Maitra for sharing his insights with our global community.