Accelerating business outcomes through digital transformation: One company's journey

As more companies reassess their business models and explore how digital value transformation can help in their evolution, one well-known brand’s journey may provide valuable insight. Mohib Yousufani, a principal at PwC who advises companies on improving growth and performance, recently discussed digital transformation at PepsiCo — including challenges faced, key improvements and benefits, and future plans — with Mattijs Backx, Senior Vice President, Strategy and Transformation, PepsiCo.

PepsiCo's transformation began when the company wanted to accelerate its digital journey across all functions. The company aimed to create an end-to-end ecosystem that could bring various digitization initiatives together. The key drivers for this transformation were the need to redeploy resources to more value-add activities, improve productivity, reimagine employee experiences, reduce costs and increase revenue uplift through digitization.

Q: Generative AI (GenAI) is one of the biggest buzzwords today. What are your thoughts?

For me, GenAI is really synonymous with more and more technologies starting to be democratized and consumerized. Many people know about them, and some have been around for decades. We just see more and more now that they’re more accessible. They're easier to deploy for companies, and that has really accelerated digital transformations.

Q: We’ve seen digital transformation drive significant value, both from a topline perspective as well as looking at employee experiences and customer experiences. How do you pull this off in large, complex organizations?

It's a journey, and it’s very important that you get all stakeholders aligned early on. Typically, in large companies, there are many digital initiatives underway, and these initiatives may overlap a bit or they don’t solve a holistic problem but more a sub-problem. One thing we did is make sure, through the right governance, that we would have the future beneficiaries — technology groups, data groups, information security — all of them sit together and really agree on both the outcome in terms of what is the digital product and what is it set out to do.

They also committed to the value that we would be able to accrue afterwards, so that we would have a clear line of sight that once we delivered this digital product, with the right change management, we can make lasting changes. And whether it's structure, whether it's roles, role redefinition, interaction models between humans ... all that needs to be part of it to truly get that lasting impact. What I found is if you get everybody aligned upfront, it's a little bit more work, but then once you're there, the actual execution time is greatly reduced.

Q: Can you talk more about point solutions? Why is that not really generating the value that companies are seeking? What do they need to do to truly increase their potential?

Let's take human resources as an example. In the past few years, if you look at the full human resources value chain from hire to retire, everyone was doubling down on a specific area. Let's assume recruiting, or let's focus on onboarding. "How do I manage laptop access, medical access? How do I start getting an ID? How do I get access to all the systems that I need for my respective unit?" Everything was a very piecemeal solution and used to have RPA embedded in it.

Today, with the advent of new technologies, especially when you bring in GenAI and some of the core components that are machine learning, you can actually remove barriers from the entire process. You can have your human resources, employees and human resources partners help with value-added functions and focus on employee growth and employee sustain versus focusing on the tactical components. That’s now fully embedded, automated and digitized, leveraging some of those core technologies.

Secondly, if you have questions that are just basic questions — for example, “Where do I find my benefit form?” or “How do I get enrolled in a specific benefit?” — I can easily leverage AI to go back and look at the forms and fill those forms out automatically versus taking time away from human resources functions. So those are the type of things that are now in play.

In finance, for example, we've looked at the whole reporting and analytics function. At one company, about 100,000-odd reports were produced annually across multiple functions. We brought in machine learning and automated the entire spectrum — which then digitized the reports — and then leveraged those technologies so that you can actually ask questions versus sifting through streams of data. “What were my sales last year versus today? Are we really growing?” Those are the types of things that are really starting to take shape and form.

Q: With all this potential, how do you advise other executives who are going through digital transformation? Where to start? How to start? What are some pitfalls they should avoid?

I think there are a couple of components. First, it’s really important that you build a very clear digital ecosystem from start to finish. From demand generation through value realization, don't cut it off. Once you deliver the product and declare victory, it's important that you loop back with the beneficiary of the automation or the digital product and confirm that the benefits have been accrued. Otherwise, you will find you have this whole stack of digital solutions, but the organization hasn’t transformed accordingly.

The second piece I found helpful is the strong partnership we have with our IT and enterprise IT technology groups. You can really leverage the tech stacks that they have built. It could be ERPs, which in some cases we were not as great at. We have so many solutions around the organization based on the fragmented businesses that we either had or that we acquired. How do you build on top of that using GenAI or machine learning?

One technology I'm personally a big fan of is conversational AI because that’s a user interface that people use to make technology work for them. I use my voice on my phone for many things. What if I say, “Hey, I changed my address,” and it recognizes me because I'm authenticated, I'm on the secure network. It then asks me for my new address and literally goes into our ERP, changes the address, creates a ticket that says Mattijs changed his address and the request was fulfilled by AI. That’s a simple example, but you can extrapolate that to other things. "I need a new computer.” “Here are the computers you're authorized to order. Which one do you want?” “I want this one.” All of that going through conversational AI.

You can build all these advanced technologies, but how you stitch them together and how you then make sure they interact with the end user in a way that truly benefits them ... I think that's a game changer.