Subscribe
Share
Share
Embed
Big Questions Answered helps us understand important CVS Health initiatives by taking a closer look at new products, powerful innovations and the big changes the company is making to achieve its strategic imperatives and build a world of health around every consumer. The company's senior leaders answer big questions from host Matt McGuire.
Matt McGuire
Let's say you're going to a concert or a football game at Gillette Stadium, and you need a snack, some lipstick or a bottle of water. Will the CVS Pharmacy that’s right next door have it even though it's getting 10 times the typical store traffic that day? “Yes,” is the short answer.
On today's episode, we'll do a deeper dive and learn how we're using machine learning to optimize the inventory at that CVS Pharmacy to ensure the right products are available at the right time. And how we're also using that process at other CVS Pharmacy locations across the country. We'll also talk about:
• How machine learning is augmenting our inventory planner colleagues’s workload.
• What prompted the company’s move towards optimizing its inventory with machine learning.
• And how both the customer and business are benefiting from this process.
Welcome to Big Questions Answered, a podcast that helps us understand the important initiatives at CVS Health. I’m Matt McGuire from the Enterprise Communications team. I’ll be your host as we take a closer look at new products, powerful innovations and the big changes we’re making to achieve our strategic imperatives and build a world of health around every consumer. Thanks for joining me today as we get our big questions answered.
I'm here today with Mario Rivera and Josh White. Mario is the senior vice president and chief supply chain and logistics officer in our Pharmacy and Consumer Wellness division. And Josh is the vice president of the inventory management and optimization.
Mario, Josh and their teams are transforming the company supply chain network to help CVS Health become America's most trusted health and wellness destination.
Mario, Josh, thanks for joining me today.
Mario Rivera
Thanks for having us, Matt. It’s good to be with you.
Josh White
Thanks for having us, Matt. Psyched to be here.
Matt McGuire
So, guys, I had to do a bit of homework to prepare for today's conversation. If I was going to be comfortable on the other side of this microphone with experts like you, I knew I'd need to learn a bit more about machine learning and inventory optimization.
So, Mario, to kick things off, when we talk about how machine learning optimizes inventory, some people may start to tune out because they don't think it really has an impact on them, but my understanding is that it does! At its most basic level, it greatly improves the customer experience for them in our CVS Pharmacy locations. Can you tell me how, from a consumer's standpoint, machine learning improves the shopping experience at CVS Pharmacy.
Mario Rivera
Happy to, Matt, and I think you got it. That’s really it. It's all about improving the consumer experience. Look, three years ago when I had the great opportunity to join CVS Health, the team was already thinking of this and starting to activate this. And the reality is that, at the time, we felt that the future of supply chain was going to revolve around an omnichannel world. The reality is that it already was at that time, and it definitely is today. What I mean by that is that our customers, the consumer, they want different options when they shop with us. They want to have the option to shop online. They want to have the option to visit our stores and have a great interaction with our phenomenal front store colleagues or pharmacists. Sometimes they want to go through the drive through and pick up their prescription that way. Sometimes they just want to buy online and then just scoot over to our store and pick it up quickly. And when they order online, very often they just want that to be a home delivery, right? They don't want to have that interaction or they're pressed for time and they just don't want to make that extra stop. So, the consumer is very savvy, and they want all of the options, right? That same consumer may shop one way one day. And then use a different channel the next day, right?
And, so, with that in mind and knowing the technology is a critical enabler for us to improve the customer experience, we started using machine learning and artificial intelligence to help us forecast future demand — so sales — to understand how much inventory we need to support our front store and our retail pharmacy. And the complexities that we have there, Matt, made it a perfect business case. For example, in our front store, we have close to 180 million product and store combinations. So, when you think about it, we have more than 7,500 front stores. I'm going to leave the pharmacy out for now just to highlight the complexity. Those 7,500-plus stores, the front store, we have a combination of up to 180 million product store combinations, so you can imagine the complexity for our phenomenal demand-planning team to manage that in Excel spreadsheets and other data and analytics tools. That's where the ML and AI comes in. It really helps us stay on top of these trends and be able to better forecast sends demand and be able to position the product where the consumer needs it when they need it.
Matt McGuire
Yeah, absolutely. I bet that goes way beyond an Excel spreadsheet like you mentioned. So, Josh, speaking of all this, there's a CVS Pharmacy in Foxborough, Massachusetts, and it is right next door to the Gillette Stadium. How does machine learning help that CVS Pharmacy and other CVS Pharmacy locations that are within walking distance of a huge stadium like that? How does it help them prepare for the audiences that attend, let's say, you know, a concert or a football game?
Josh White
Yeah. Well, believe it or not, we have begun integrating hyperlocal weather trends and local event data into the logic we use to forecast our demands by leveraging machine learning and AI, we can bring together historical data around like very specific types of events, like football games that you mentioned, concerts, and we can automatically understand what demand those specific events drove in the specific geographies and by store and by category. And then we can use that to basically dynamically sense or predict the impact of similar upcoming events on future sales and other places. Like, for example, you know, you were sort of getting there, but like we can understand how much red lipstick will we need for a specific store when a specific major tour comes to town. And this, you know, this approach really has helped us strategically position inventory to maximize those sales during, you know, these high traffic events, but also, it helps us minimize the risk of overstocking after the event is over.
Matt McGuire
Yeah. And, so, you know, just to kind of follow up a little bit on my first question, too, where we were talking about, like, the consumer experience, what are their business benefits of machine learning? What does that offer as it optimizes inventory at a CVS Pharmacy location?
Josh White
That’s a great question. So, well, obviously if we're able to better understand when we need red lipstick in a store ahead of a tour of that, then that means we're going to be able to capture more sales because we're more in stock than we otherwise would have been. But it also allows us to send less inventory to stores. And it may seem counterintuitive to simultaneously have less inventory but also better in-stock — you know, better in-stock experience for customers. But improved demand forecast allows us to reduce our safety stock. So, safety stock, for those who you know aren't inventory people, is the inventory we carry as a buffer to guard against unanticipated fluctuations in demand. So, if we’re able to anticipate demand by leveraging ML and AI to improve our demand forecasting, then we can improve in-stock, sure, we can drive sales, yes, but we can also reduce the working capital we have invested in inventory, so we can use that money to fund other areas of the business.
Matt McGuire
Yeah. And obviously, when both you and Mario are saying ML, AI, machine learning, artificial intelligence. My understanding is that, like, machine learning is sort of a subset of artificial intelligence. Is that correct?
Josh White
That's right, yes. Machine learning has been around, but AI — the capability to bring together all of the vast amounts of data that is available to us in one form or another externally, and then use the self-learning, machine-learning algorithm that sits on top of that data to generate this much more enhanced, much more sophisticated understanding of what demand looks like in the future, that’s new.
Matt McGuire
Got it, got it. So, Mario, at CVS Health, we're always looking for ways to innovate and update our processes, and as we get tremendous help from our colleagues in the digital, data, analytics and technology department to incorporate the latest and greatest, what prompted this move towards using machine learning to optimize our inventory?
Mario Rivera
Great question, Matt, and as I said before, and some of the points that Josh was just making, you know, we came out of the pandemic, and like many other companies and retailers, we were struggling. And the effect of the pandemic was pretty harsh in terms of our in-stocks, our product availability on the shelf — very visible, particularly in the front store. And then combined with the fact that we had a very talented team — we still have a very talented team, you know, in our inventory management organization — they were doing all of this planning very manually. And so, you know, the mindset here is: How do you help your team get better with improved processes, and how do you layer technology on top of that to make it happen? And so, we had a perfect combination here where the business was struggling, our out-of-stock conditions were pretty bad, we had a great team in place with a good process, as manual as it was, we had a good process that was ready for us to layer technology on top of it, and, as I said earlier, with a high degree of complexity right up to 180 million product store combinations just in our front store alone, right? I’m not even getting into the complexities of the of the pharmacy. And so, it was very clear for the team that we had an opportunity to improve our colleague experience, and by doing so, we would be able to improve the customer experience by improving our in-stocks, our on-shelf availability of our products in the front store, and at the same time, reduce nonproductive inventory, excess inventory, that we clearly did not need to service the customer. And so it's been a fantastic thing. We've seen so much improvement and sustained performance in our in-stocks and on-shelf availability for the last two-plus years now. And we've been able to shave off a ton of inventory from our balance sheet. So, super proud of Josh and the team, they’ve been doing a phenomenal job.
Matt McGuire
Hmm, interesting. So, Josh, I mean, based on kind of what Mario just outlined, I imagine employees in the CVS Pharmacy locations are seeing plenty of smiles from customers based on this improved inventory. But how is machine learning impacting our inventory planners?
Josh White
That’s a great question. Yeah, I mean, as you can imagine, it’s a big deal in the life and the work of an inventory planner. But you sort of think of these capabilities, this machine learning and AI capability as a tool in the toolkit of an inventory planner. It's something that augments their work, but, basically, it’s a really powerful tool they have at their disposal. It allows them to spend less time, you know, manually manipulating or updating the 180 million as my image and you know item-store level forecast that we maintain, and instead they can spend more time on more strategic elements of their job like partnering with suppliers to ensure we have the right supply in the upstream supply chain or focusing on what we call product-lifecycle management, which is basically like how do we introduce new items and get out of old items in the most profitable way. And you know what's remarkable about the tool actually is or the capability is how our inventory planners have embraced it. I think that it's partially due to how we approach implementation. We didn't roll it out in a day and say, like, you know, this is the new capability, you know, for you guys, use it. It was, on the contrary, they worked side by side with our tech partners and our digital partners to help define, you know, what it needed to be, you know, what the requirements were, and our planners were key to the testing and iterating on capabilities as we scaled it. They were very much part of driving its adoption. They very much see the power of it and are now, you know, utilizing for almost all the man forecasting we do as a result.
Matt McGuire
Yeah, it sounds like it's doing a really nice job. So, to close things out, Mario, what would you say you're looking forward to most about how artificial intelligence is helping us within Pharmacy and Consumer Wellness?
Mario Rivera
Wow, now you're getting me excited. I could talk for hours about that, Matt. Look, I'm an engineer. My background is in engineering. I've had the pleasure of leading engineering teams for most of my career. And I I think of myself as someone that tries to keep up with technology. So I've been a user of AI and many other technologies for some time now. And I really think that unlike other tech bubbles, I don't think that this one is going to burst. I think AI, gen AI and everything else that we're seeing is here to stay and we're going to see increased power and use cases to adopt it. So, I'm super excited of how we can continue to use AI to augment the workload and how we will improve the colleague experience, as I mentioned in the example that we have live now here in our supply chain logistics team. But also how we continue to better serve the consumer and our patients. And we have work underway to pilot other things. Have another pilot that is getting started in in Josh’s prior role in transportation and logistics, where we're going to do some routing optimization with some AI support to better route our trucks or trailer or deliveries to our stores so that we can do it in a more timely and efficient manner. And, this summer, we also had a pilot to try out AI for safety, basically a tool that allows us to detect and predict injuries and other ergonomic concerns in our DC operations so that we can make better decisions and how to improve the colleague experience and make their job safer and easier as well. So, sky’s the limit. There’s a number of use cases across the end-to-end supply chain spectrum, and my team and I are committed to continue to explore those use cases and partner with our DDAT team to bring them to life.
Matt McGuire
Wow, those all sound pretty interesting. Let’s stay in touch. I would love to find out more about them a little bit down the line.
Mario Rivera
That would be awesome. We would be happy to do that, Matt.
Matt McGuire
Mario, Josh, this has been a fascinating conversation. I gotta say, a lot more interesting than the homework I did to prepare for this call. Thanks, guys, for joining me today.
Mario Rivera
Thank you. Thanks for having us.
Josh White
Yeah. Thanks, Matt. Look forward to talking to you again about those use cases Mario's mentioning. And I just want to say, we are really proud of all the work the team has done in this space. It is a total game changer like Mario said. And we'll keep you posted on how it evolves.
Matt McGuire
Absolutely. Thank you very much. And a big thanks to you for tuning into this episode. Until next time, I'm Matt McGuire. I look forward to joining you again to get more big questions answered.