Todd Foley: Hello, and welcome to the CDO Magazine interview series. I’m Todd Foley, CDO and CISO with Lydonia. I’m joined today by Ellen Webb, Vice President of Shopper Analytics and Insights at PepsiCo. Ellen, thank you for joining us today.
Ellen Webb: Oh, happy to be here.
Todd Foley: When you look at how you’re leveraging data analytics to optimize customer interactions across multiple channels—right, at the same time, with the same person potentially engaging both in their physical life and their digital experiences—what recommendations would you give to people about how to approach that? What are the principles you use to guide that? It’s a tough road for anyone in this space.
Ellen Webb: Yeah. And I would say, as we think about the future, there are some basics that shoppers expect, regardless of the conditions. We’ve recently done some work, again, through our Pepviz approach, as we think about the fidgetal spaces. I think there are general principles about how people can think about driving growth into the future. And again, there’s a range of speed at which this could happen and the depth of impact. But I would say the first one is convenience. This has been top of mind for as long as I’ve been in this industry. But I would say there’s a bit of a shift as we think about the future, which is the concept of autonomous convenience. So, it’s really shifting from thinking about ease and speed to that, plus the ability to offload the mental labor associated with shopping. So, if you think about having to create a list or being able to find everything that I need in one place, no matter what kind of format I’m shopping in, and just having that flexibility—that’s something we’ve found is absolutely critical to winning with shoppers in the future. And the fidgetal concept plays a big role in that. The second one is really about loyalty. We talked a bit about loyalty earlier in the conversation. Especially in today’s environment, shoppers really want to feel appreciated. They want to feel understood. And they want to feel that whoever they’re buying their groceries from values them as a person and values their loyalty. That comes to life with product recommendations, specific offers, and promotions that are relevant to them. That concept of precision loyalty—saying, “I’m choosing to shop at your store and buy your brands,” and making sure that you’re rewarding them on the back end for that. So, I would say that’s the second piece. The third piece, which I think ties in with both of those, is assortment. A lot of what we do comes down to what products you’re offering to the shopper at that point of purchase. That space is so complex, with so many interconnecting sets of needs. For a shopper, we talk about anticipatory assortment—how are you anticipating what they’ll be looking for when they get to the store? Do you have the products that fit the wellness needs I’m looking for, or affordability, or fun and enjoyment? There’s a lot that goes into those specific points of purchase and the mindset people are in when they’re there. It’s not just helpful; it’s absolutely critical to winning with shoppers in the future. So, I’d say those three things are what we’ve honed in on. I think technology helps in each of these areas, but it also extends into the physical store as well.
Todd Foley: I think being able to give people positive rewards around their association with the product will always drive brand loyalty. We talk about those three things with two things that are a little opposed to each other. That autonomous convenience, right? Not having to think about it at all, is the promise there. Then assortment—being able to give people myriad choices for what they might do. That’s a difficult tightrope, I would think. You want everything there in terms of it being like an unlimited menu of options, or at least an appropriate one, and not have them think too hard at the same time.
Ellen Webb: Yeah. And I think it’s not an endless set of choices. There’s lots of research showing that too much choice actually feels like you have less choice because it becomes overwhelming. I think that’s what we try to get into. What’s the right assortment that meets the bulk of needs? It’s that balance between having everything but having the right things. Especially when you think about balancing, that’s the other part of this that comes in—balancing the shopper needs with the operational realities of fulfilling that, too. Those are the things we sort of take into account. As we talk about that, we’ve got a lot of examples where we’ve worked with retailers to optimize their assortment in-store relative to the shopper and demand that is in that specific store. And we know it works, and sometimes you have to make difficult decisions in terms of what goes in and what comes out. There’s a lot of science and analytics that goes into that work.
Todd Foley: Yeah, it’s an interesting challenge. I’m sure what you identify is relevant variables are still finite, but it’s way more than two, from what sounds like.
Ellen Webb: Yes, it is. Way more than two, for sure.
Todd Foley: As you look at how you’re pulling data together and as you’re trying to not have a crystal ball but directly impact some of those things in a measurable way, there’s a challenge there. A constant challenge for you and your teams around being able to refine that data, get greater accuracy, and have greater impact. And it’s consistently engaging with the business and retailers in a way that is a little bit different from some people in data roles, where, while they’re involved in that process and that value chain, if you will, they’re able to produce insight and impact the business. You’re playing a really direct role. I think that’s a remarkable tribute to what you and your team do, and being able to do that, but it also must give you a unique perspective on the business. Are you routinely engaged for things that might surprise you? Or can you share some examples of where some of these fidgetal approaches have played out at PepsiCo?
Ellen Webb: Sure. And I will say the way our team is organized is we do a lot of the work and help to enable the balance of the organization to do this. So, a lot of our category management partners, sales strategy—they’re our critical partners in delivering a lot of this to the business. We’ve had tons of success, and I think different development from our analytics team in being involved in those conversations, which I think is critical. So, just a specific example. If you think about, again thinking about the concept to solve, this happens across the business on a fairly regular basis. We have a retail partner whose got shoppers with a lot of different business needs. If you think about a convenience and gas store, they can have thousands of stores, and even within a particular zip code, you might have two stores that are less than a mile apart but have very different consumer bases, shopping behaviors, different needs, and different demand profiles. And so, as you think about that, it’s really important to say, okay, what are the things that we need to do to balance, again, that kind of the concept of anticipatory assortment, but also paired with, again, how do we operationally do this? So, it’s not, you know, you don’t have hundreds of thousands of unique planograms across the organization. So, in this situation, we realized, okay, what are the things that we can do to ensure that we’ve got the right assortment in the right store to help them grow dollar share within the market? So, again, growth being the goal. What we did is really worked with them again through a pet beds approach and said, okay, we want to drive an advantage category strategy and planogram strategy to really deliver against what those shopper needs are, with the goal of driving sales, profit, and market share, again, making sure that the shopper is central to that. So, what we did is leverage our store DNA solution to really understand who the shopper is in each of these stores and what they’re buying, both when they’re inside and outside of the store. So, looking at that at a store level and then clustering and segmenting that up into different targets or basically clusters that help to, again, say, okay, we’re going to be practical in terms of how many clusters make sense statistically with what’s viable operationally. And then targets had to execute the assortment. So, I mentioned store DNA earlier. That’s a key data set that we have that combines data from, you know, we’ve got data on over 500,000 retail stores. It helps us understand what the stores sell, what the attributes of those stores are, and what happens around those stores for different points of interest that help us create a profile and say, okay, here’s the profile of each store. And again, how do those start to cluster together? It’s a really powerful data set that we have, that we can, you know, point into commercial solutions. So, in this example, based on the demand dynamics and the shopper dynamics, we landed on four unique clusters. One focused on healthier items, one focused on functional beverages, another focused on flavors, and the last one was just kind of your core mainstream beverages. Once those clusters were determined, then the team created the planograms to reflect, again, the categories and brands that over-index within those clusters. This was in 2023. The single-serve beverages, so your 20-ounce on-the-go beverages, grew about 9 percent in the stores that were clustered, compared to about 6 percent growth in the stores that were not clustered. So, a pretty significant difference between those two. Then, year to date in 2024, those trends continue. The clustered stores are growing about 5 percent, whereas the stores that were not clustered are about flat. This is again talking about ranges. We consistently see really strong growth when we do this. So, it’s something that we continue to work with our customer partners to initiate.
Todd Foley: There’s almost immediate gratification there, where you’re able to track the real material impact of the work you do and how it’s being leveraged out in the field, and that gives you a constant feedback loop to always be tweaking and improving and looking to take other approaches. But it sounds like you’ve got an embarrassment of riches. You talked about a dynamic where you were trying to limit the number of clusters you were applying just from a feasibility standpoint. In theory, you could create half a million personalized…
Ellen Webb: Sure, we can create all the versions in the world, but we have to really be, which that’s part of the benefit of sitting in the organization that we do, very conscious of the operational complexities that we don’t want to create a nightmare for our field and our customers to execute. And so, there are lots of those operational considerations that come into play as we do this to make sure, again, that we’re balancing, what’s the right thing from a shopper perspective? What’s the right thing to grow the category? But then making sure that we can execute it at scale is just as important.
Todd Foley: And you have the ability to measure those results in such a quick way that you’re never asking for too much, right? That’s, I think, a fantastic position to be in. It’s something that we always like to talk about, the connection between insight and impact, in being able to improve that time to impact, not just improve the time to data. You have that in a very concrete way that a lot of people would be jealous of. You have that ability and you’ve had that impact on the business in such a measurable way. Does that create additional engagement because you’re able to measure it that way? Because you make such a difference in the business, do you then get asked for more and more?
Ellen Webb: Yeah, of course. And here’s the other thing that I think is important to say: as we do this, it’s not a one-size-fits-all. So, in this situation, we landed on four clusters, and I talked through what that was, but for a different chain, it might look different. So, we really have to tailor the solutions based on the specific needs and problems to solve of those different customers that we have. But the other piece is, I talked about the success that we have, but a lot of this has come on the heels of tons of tests and learning. So, there are things sometimes we try, and we may not get the results that I talked through on that one. And so, it’s then going through the postmortem and saying, okay, what do we think was, you know, what went wrong? What are the things that we need to do moving forward? So, there’s, again, that I think being able to learn together and say, okay, here are the things we agree that this is the right thing to try. Let’s make sure that we test it. And, in that example, we had the ability to look at clustered stores versus non-clustered stores. We don’t always have that. And so, the more that we try to do to, as we’ve tried to further this agenda, it’s about collecting those proof points to say, like, hey, we’re not just making this stuff up. We have reason to believe that it works. And I think that’s the other piece of this is, we can’t just go in and do whatever we want in these stores. It has to be done together with our retailer partners. And we have to have the credibility and the proof points to get them aligned to doing that. So, which I think, at the end of the day, makes us better because we’re really being thoughtful about the things that we believe in. And also, being honest where we think that we’ve got potential, here’s a potential concern or things that we might that might keep us from enjoying the success that we want to in this. And so, again, really thinking about with them, what’s required end to end all the way through execution and making those required trade-offs. And so, it’s, I think it’s, as we’ve gone through this, and all the things that you said are absolutely true, we’ve got a ton of great, you know, resources at our fingertips to be able to do this, but we have to be humble, and there were just a ton of lessons that we’ve learned in the process. And, in the business, we have to be agile because the business changes every day, and the world that we’re operating in is fickle. And so, that ability to be agile and responsive to what those needs are is critically important.
Todd Foley: Yeah. And you have a need to create that same trust with your business users, retailers, and partners, just as we were talking about that consumer trust. You have to be able to move the needle on. That humility and accountability that you and your team assume, it is key to making sure that your guidance is sought after and embraced instead of it being something that’s seen as a challenge or having lower value. I think that there’s a virtuous cycle there. That your team has to be involved in at all points, right? Not just being able to produce the insight or make recommendations, but also be accountable for that and be able to adjust. And I think that level of engagement with the business is truly pretty unique. I think you’ve created a rare value prop that’s immediate in terms of business engagement and responsibility for impacting the business. That’s good. That’s got to mean that you’re asked to do more and more all the time, though.
Ellen Webb: Sure. Yeah, it is. But I think it’s like anything, even if we had endless resources, just because we can do it doesn’t mean we should. We do a lot of prioritization, and that’s a lot of what we’re trying to do is how much more can we automate through technology in terms of a lot of this stuff. So that we’re at the top of our conversation, I talked about the process of data engineering and data cleansing and management. So, the more that we can offload those types of work we’re automating to make that less time so that we can free up the energy and the time for the value-driving activities, but you’re spot on. Like we do absolutely have to make trade-offs, and it happens every single day. But it’s one of those things where we wish we could take on all the projects, but at some point, we reach the end of our capacity. So, it’s a constant trade-off, and making, again, those decisions on where we think that we can add the most value, but also helping to enable the teams across the organization to, again, democratize analytics and, have them enabled to do the work as well, and certainly be able to talk about it with their customers. I think a lot of times analytics feels like it’s this thing that needs to be, I think people sometimes are scared of it. And that’s a lot of times what we’re trying to do is just making it approachable. And it’s something that people can understand, and really making, again, we want people to want to make analytics-driven decisions and impact the business in that way. And so that’s part of it too, is, we don’t necessarily own all of it. There’s, again, we’ve got an army of people who are out taking these capabilities out to our customers as well, which is a great part of the design that we have as well.
Todd Foley: You make such an immediate impact on the business with what you do, and you’re involved in both a retailer partner and customer engagement standpoint in such an immediate way. You said something there that I think is really critical: even as there’s greater demand for your team’s time and focus, you’re trying to leverage technology to take some of the time-consuming drudgery out of what they do to allow them to be more innovative and focus on greater demand. But you’re also trying to bring it closer to the business and democratize analytics in a way that truly allows you to address that demand. I think self-service data has been a term that’s been around so long, I’m not sure it means anything anymore. How do you approach that democratization of analytics at PepsiCo?
Ellen Webb: Yeah, I would say it’s through, I’d say a lot of it, again, we’ve got partners to help us do that. It’s about—it’s just as important to have people trust in the analytics, but also, again, to my comment on making it approachable. What is the user interface to the point where people don’t realize they’re being an analytics person when they’re poking through a visualization or a dashboard, and they’re like, “Oh, this is amazing. I can’t wait to take this to my customer.” And so it’s really about that user experience, which we’ve found to be super important. And, at no point today, I talked to you about the different kinds of data sets that we have, but I haven’t talked about how many petabytes of data we have, because that doesn’t really matter. It’s what you’re actually doing with it. And so, I think, we’ve got a lot of partnerships with some of our digital engagement teams or digital transformation teams that are working to bring a lot of this stuff to our frontline. They might not even know that it’s driven by a massive analytics model, but they know that they’ve got information at their fingertips that they can use. And so, again, it’s really about simplifying, making it approachable, and making people want to use analytics and seeing it work for them. And I think that whole model is something that we’ve found to be very, very impactful, instead of several, probably, I’m not sure how many years ago it was, but I think we were in a “if you build it, they will come” type of environment. You’re building this big, fancy thing, but people are like, “I don’t know what to do with it.” Versus now, the model that we’re really thinking about, again, through this pet-vis approach, is listening for what those questions are and how we bring those into solutions that people can actually use on a day-to-day basis. And so, I think it’s also this mindset of little wins being important too. When they start to pile up, they can become big wins. So, making sure that, again, you’re starting to get those proof points to say, okay, this is not just a little idea. It’s a big idea, and we should do something with it. I think that concept of approachability and democratization— we’re not giving everybody full access to all the data that we have, because that would be a bad idea on multiple levels. But again, being curated for the things that matter and getting it into a way where people just see it as something that is a natural part of how they run their business versus something that somebody is telling them they have to do.
Todd Foley: Yeah, I think it’s much more impactful that way. Truly remarkable. And it’s been a pleasure hearing about what you’ve done at PepsiCo and how you’re approaching it. As I said, I think you’re pretty far on the scale in terms of where people would like to go. You’re there in a lot of ways, though, as you said, that comes with its own challenges, such as demand and constant engagement and being able to combine both the physical and the digital worlds in a way that’s immediate and impactful. Before we wrap up, is there any guidance you’d give to others in your position, maybe who haven’t reached the evolution that PepsiCo has? Any lessons you’ve learned along the way that you want to share?
Ellen Webb: Yeah, I think that’s a great question. I would say again, the first one is just really listening to your business partners and focusing on the things that you can drive that have more immediate impacts to build trust and create value for what you’re trying to do. Because I think part of having been in a category management role or just as part of the bigger sales organization is, I’m an insights person, but I’m always selling. It’s a very objective perspective, but I’m selling the validity of the analytics that the team is doing or selling the perspective of somebody to care about the shopper. But again, it’s being empathetic and understanding of what is it that matters to the person on the other side of the desk or the computer screen. And so that you’re bringing things that are relevant to them. And I would say the second thing is, you have to have some level of humility and say there are a lot of things that I don’t know about what they’re trying to solve. And being able to come to a collaboration saying, “Okay, here’s the value that I think I can bring to a conversation versus what somebody else can bring.” And really being super clear about that. And I think throughout our conversation this afternoon, there have been so many things where I’ve talked about all the cross-functional partners that we have. We could not have the impact that we have or do what we do without our cross-functional partners. So, whether it’s category management, sales, IT, strategy, or transformation, it takes a village to get this stuff done. I think being able to build that coalition, again, focusing on what are the things that you want to do, is super important. And then I think the last thing is just choosing, we talked about prioritization, choose one or two things and do them really well. And, again, use that to build trust because I think sometimes things fall flat when you try to do 15 things and fail at all of them, instead of focusing your efforts on a couple of places and getting those wins. So, I would say those are probably the three biggest learnings. And then the last one is just be patient. These things take time, and sometimes it feels like a frustrating journey, but once you start to see it, things pick up, and it’s a really exciting thing to see the impact of the work really take place, which is very fulfilling.
Todd Foley: That’s great guidance, and Ellen, thank you so much for joining me today. For those of you listening, please visit cdomagazine.tech for additional interviews. But Ellen, it’s truly been a pleasure. Thank you.
Ellen Webb: Thank you so much.
