Ep 56: Orchestrating the Global Warehouse through AI with DHL Supply Chain Global CIO Sally Miller

Evan Reiser: Hi there and welcome to enterprise AI innovators, a show where top technology executives share how AI is transforming the enterprise. In each episode, guests uncover the real world applications of AI for improving products and optimizing operations to redefine the customer experience. 

I'm Evan Reiser, the founder and CEO of Abnormal AI.

Saam Motamedi: And I'm Saam Motamedi, a general partner at Greylock Partners.

Evan Reiser: Today on the show, we bring you a conversation with Sally Miller, Global Chief Information Officer at DHL Supply Chain. DHL is a global logistics powerhouse operating more than 2500 sites worldwide, with billions in revenue and over 60,000 employees. The scale is staggering. The company's literally moving the world's goods every single day. 

There are three interesting things that stood out to me in our conversation.

First, DHL has deployed over 8000 robots across its facilities, working side by side with people to reduce strain, increase throughput and even improve job satisfaction. A great example machines and humans working together. 

Second, we talked about computer vision solutions like trailer unloading bots and driver monitoring systems are making warehouses safer, while multilingual robot interfaces are creating more inclusive workplaces.

Finally, Sally shared powerful examples AI at work behind the scenes for machine learning models that predict demand. Orchestrate daily operations for generative AI agents that confirm delivery promised hours instead of days. The result faster, safer, and more efficient operations at a massive scale.

Evan Reiser: I first of all, uh thank you. Thank you so much for joining the show today. We were looking forward to having you. Uh do you mind giving our audience a little background about yourself and, you know, the role you have today and maybe I got there. 

Sally Miller: My name is Sally Miller, and I'm the global CIO for DHL Supply Chain.

DHL is, you know, made up of five divisions. The division I'm in, we provide third party warehousing and transportation services to customers. But a lot of us might know us from our express division, which is time-definite parcel delivery, as well as our forwarding division, where we move freight between continents. And we have a division called P and P, where in Germany the post office is privatized and we basically run the German post office.

And then the last division is a division called DHL E-commerce, which inserts packages less than five pounds going to consumers and to, you know, the Postal Service here in the US and other countries and to their, network to provide, reduced costs to typically e-commerce shippers. So I've been in it my whole career, which is over 30 years.

I started out as a COBOL programmer, and I'm sure a lot of your listeners don't even know what that is. 

Evan Reiser: So some do. I've got a couple of…

Sally Miller: Good, good. It's, you know, I think was a good base for IT to learn application development. And, I got into supply chain IT fairly early in my career after doing five years in banking and have have just progressed from programmer to leading projects to, managing teams, developers as well as implementers.

I've been with DHL now for 20 years, and, I really have seen the role of the CIO change during my tenure as a regional CIO and a global CIO. It's less about uptime and key KPIs, and now it's all about innovation. Whether that be robotics, data analytics, AI, it's how can we innovate, make our operators more efficient and provide the tools that, you know, white collar workers and even blue collar workers expect.

You know, we do see that, associates are much happier working in an environment where technology is deployed. So supply chain is historically known as being very manual, a bit dated. Well, things have changed significantly probably in the last ten years. 

Evan: DHL, an interesting company where I think it's, you know, the the brand is pretty well known worldwide.

Sally: It is.

Evan: I'm sure. I'm sure you haven't heard of it, but but I suspect that, like, very few people truly appreciate the ethic, the scale of operations and maybe the impact that you guys have of the world. So is there some way to help contextualize that, like, you know, how would you how would you, maybe share with the average person, like to help them really appreciate the the full scale impact of, DHL Supply Chain?

Sally: I think that a lot of people in the US are very familiar with UPS and FedEx, and DHL does not do domestic parcel delivery in the US. But if you go outside the US, Canada, Europe, Asia, Latin America, the brand is even more well known than it is in the US. For supply chain, the division I’m in, we operate 2500 sites globally and we're the, the industry leader, and we're larger by double our our next competitor.

So, you know, bigger isn't always better, but it is important, you know, in today's world, a global economy to be able to provide services for customers in all geographies. So that is a big differentiator for us. And I think that overall we're about the, you know, top ten in terms of the largest employers in the world, private employers, you know, not including governments, but, it's a very large organization, but very, you know, great opportunities.

You can, you know, change roles without changing companies. And this supply chain is where your heart is and interests. It's it's been great for me personally. And, I've enjoyed helping others achieve their career ambitions as well. 

Evan: We think about the kind of the last like, you know, 10 or 20 years, you know, what are some of the like, more exciting kind of, or impactful changes you see for a technology perspective? Right? Across the last couple, you know, technology platform shifts.

Sally: There are a lot of technologies that are hyped. But I think that even the ones that are overhyped, there is something you can get out of them and deploy in your business. But one of the challenges that we have and a lot of companies have is you really need to have good quality data to leverage some of these technologies quickly and at scale.

Otherwise you have to, you know, take it in bite-sized chunks, which isn't tragic. It just takes longer to realize the benefits. But probably the biggest, uh, advancements fall into a couple categories. The first would be robotics, and we have over 8000 robots deployed across our 2500 facilities. And these truly are, solutions that work alongside people to make their jobs less manual, less strenuous, and more enjoyable.

It reduces training time. They're more productive. We get more throughput, more or less dependent on filling some positions that, you know, have remained open even post-COVID. But things like unloading cartons out of a hot trailer in the summer, you know, if you've got a bot that can do that, bots that, you know, help people pick orders by moving the items from one zone to another so that there's less travel time for our associates who are picking orders.

And, you know, then delivering the completed orders to the packing area. So these advancements have have really made working in these environments just a lot more pleasant. You know, another, key technology is, you know, different forms of AI has been around for over twenty years. But machine learning, there are a lot of use cases where we can take data from different applications as well as from external sources, and learn and predict what will happen so that we can make decisions and influence outcomes better than we've been able to in the past.

And when you have a company like ourselves where you have 2500 sites across the world, you can't really, you're not physically there. And just to get the data feeds, it's helpful to combine that and have the algorithms learn. It just saves time and provides insights and makes us more efficient. So machine learning and then we had, you know, I come out, ChatGPT made a big splash.

And it was how can we leverage that in some of our back office functions and our ability to side use cases where we have good data and have the model learn from the data and, and produce what is being done manually is a big run for large organizations like ourselves, where you have people, you know, responding to RFPs or responding to associate claims.

There are a bunch of different use cases where we have years of data and years of responses that we can use AI to, to do that type of work alongside, you know, a human in the loop in terms of reviewing the output, just saving a lot of the manual, redundant, boring work makes the roles we're able to offer, a lot more interesting.

And, and then obviously, lately, agentic AI has gained, a lot of hype. But, I really am excited about some of the use cases that we have that really takes manual labor and really lower cost labor markets, where we're using call center support in Costa Rica or India. We can have an agent do that work. And, the agent is able to complete the work much faster than humans.

So where we would be calling, you know, over the course of a week to confirm delivery appointments, we now kickoff agents and those, you know, appointments are confirmed within a couple of hours. So we've saved that manual. Some people call it soul-sucking work. We've automated it and we have the data faster. So the downstream processes are more efficient and we have more opportunities.

So those, those are the areas that are hyped. Now, there are some things that are way overhyped, one of which I would say are humanoid robots. You'll see a lot of videos of robots that look like humans folding laundry or, you know, running down the street or dancing. But the use case in supply chain for those so far is pretty limited.

They have limited battery life and the cost there's really not a return, but they, they do get a lot of visibility, hype and investment, you know. So, I do spend a lot of time talking with venture capital companies and where they're investing and trying to persuade them, you know, to to fund companies that are developing products that are fit for their intended use, look like what they're intended to be used for?Not necessarily, a scary looking Star Wars type figure. No one wants to work next to that. It's scary looking. 

Evan: You guys, I'm sure, should been doing machine learning. You know, your whole industry has been like when the leaders right now actually machine learning from way back, even like thirty, forty years ago. Right. There's a lot of, you know, students in school studying computer science. They probably didn't realize that some of those problems, the key computer science problems come from you know, logistics and the supply chain. 

I'd love to ask you, like, where do you think we are in that that hype cycle with AI? 

Sally: I think we're early in the hype cycle, but I don't think you can compare the hype cycle of AI to other technologies in the past, because I do think the potential impact I view AI is to the white collar workforce, what robotics was to the blue collar workforce. It is going to change the way people work, much like the introduction of the personal computer. I mean, if you look back at films from the 60s, you saw those large typing rooms with, with people sitting in there typing all day, and then you, you know, went to the early 80s and people were on, you know, laptops. I think that AI is going to have that large of an impact on how people work in the white collar environment.

So I think the hype is justified, and I think that the, the costs to deliver AI solutions and the need for good quality data and the, you know, the ability of solutions to reason and act independently will improve over time. And I think all companies need to be looking at use cases where people are spending a fair portion of their day doing non-value-added tasks and how can we use AI to do those tasks for them? 

Evan: I appreciate your perspective, Sally. So I you know, my background is not in the physical world. Technology. I work in cyber security. I have to imagine if I went to a DHL facility, I'd probably be blown away by, I think just the, the level of the, the technology deployed.

I have to imagine most of our listeners, even outside technology, would kind of be blown away. 

Do you mind sharing maybe a couple of ways you guys are using kind of AI or other kind of transformative innovation that these that the average person wouldn't fully appreciate, right. Even though they might be picking up a DHL package once a week.

Sally: Well, I think that when you enter a DHL facility, you can't necessarily see all the technology deployed because, a lot of it is looking at, you know, what the order pool is for the day, what the available labor is, and matching the most efficient way to, pick, pack and ship those orders to meet the service level and the available resources, whether that be people, automation, robots, whatever form.

So there is a lot of orchestration in the background that wouldn't necessarily be visible right away, but that in itself is fairly new to the industry because that would be done manually. Previously, someone would, you know, kick off. You know, we call them tasks for people to complete.

And and now we can automate that with intelligence to optimize that instead of it being tribal knowledge. You do this at ten o’clock. You do this at twelve o’clock. So you know, a fair bit isn't very visible to the AI, obviously, robots are. So seeing robots, and we allow the associates to name the robots to make it more personal.

So, you know, little things like that are things we've learned over time to help with change management and make it a fun place to work. But, I think that managing 2500 facilities, our number one priority is, is safety. Ensuring that people leave the same way that they arrive to work. So safety, physical safety of the environment and people is priority number one.

And you mentioned cybersecurity. Yes. You've heard in the past year of several, you know, threat actors targeting customers, supply chain as well as supply chain companies. So the amount of investment we've had in defending and shoring up the environment, as well as the, you know, auditing the environment of some of our key software suppliers to make sure that they don't have any exposed vulnerabilities so that we can ensure a smooth level of service not only of applications we host, but ones that we utilize every day that are our mission critical.

So, do keep in mind that, you know, our express division is, is moving parcels and letters and mainly is a network of businesses that have sortation centers and very tightly tied to the inbound and outbound schedule. And they are architected for peak volumes for meeting, you know, the predicted volume, service levels and just see the infrastructure and level of automation of those facilities is very impressive.

On the supply chain side, we typically are providing warehousing and transport value added services for companies who don't want to execute that portion of the supply chain and leave it up to the experts that, this is all we do and this is all we do. So mastering the deployment and use of supply chain IT applications, warehouse management systems, transport, labor management, there are a bunch of them, and knowing which robotic solutions to deploy based on profile having that integration being able to maximize the efficiency of not only that robotic solution, but of the humans interacting with it, is a lot of what is happening behind the scenes that you wouldn't see when you physically walk in.

But it is all about optimization. People, automation, robotics, and the software that houses the work that we need to get done that day. So, I've managed of a team of over 3000 people globally, and it takes a lot to keep up with a fast growing company like we are. We have ambitions to double by 2030. So that is our our mission and we've got plans. 

Part of it is growing our current customer base and new customers. But also we've done some acquisitions and we are moving fast. 

Evan: So I think if you kind of reflect on the last maybe couple of years, are there any are there any kind of innovations using kind of AI or robotics?

Those are those are becoming increasingly, I think, the same thing. Innovations you feel particularly proud of where you maybe, you know, maybe a former version of yourself? Right. Sally, from ten years ago, we said, hey, that problem, that's just the nature of our business. There's, you know, we just have to kind of deal with that. And now in hindsight, like, wow, like with the with the kind of innovation these new technologies really, really kind of reimagine what's possible. Right? And yeah, kind of led to a, a big win that feels, you know, memorable and, you know, memorable and, innovative.

Sally: I would say some of the solutions that we've deployed that leverage computer vision have really made me sit there and say, wow, you know, I mentioned unloading trailers. You know, we've got, a bot deployed from Boston Dynamics that is looking at the size of the curtains and determining which one to pick next.

And if one falls off, it's smart enough to, to pick it up. And just that it is is very impressive. We've had, another use where we monitor driver behavior in our trucks. So seeing if a driver is, you know, nodding off to sleep or if they're on their cell phone or if they don't have a seatbelt on.

Just improving the safety of our drivers is very important, and it's helping them. And we will see films where they were looking the other way, and we will alert them and they'll give us a thumbs up like a thank you. You know, they were distracted and they're they're back on track. So that type of interaction of technology that we've deployed, it is rewarding.

And probably the third is the bots in the warehouse that are, you know, able to display and any language that's selected by the associate. And just seeing how happy they are. You know, if you're an immigrant from Somalia and you're in a warehouse in Columbus, Ohio, to see your native language makes you feel at home. So, I, I do like to see the bots named after something that's familiar, to them and the native language.

It makes people feel part of the team. And, much more inclusive environment, which I enjoy being a part of. There are a lot of things, if I look back ten years ago of what I'm doing today, and probably one of the most rewarding is that we are more, you know, a respected and valued member of the board and not just a cost or a necessary evil, but we are leading the business.

And how we operate and how effectively we do that is, you know, how successful we are in our roles as CIOs. 

Evan: Way back in, the early days of internet, I remember buying computer parts online as my my hobby as a kid, and I would like send a check to the internet to then maybe, if you're lucky, something would show up, right?

Sally: Right. 

Evan: In the future. Today I feel so spoiled, right? If I don't get a text alert where the truck is, four stops away and I know when exactly they press my doorbell, like I'm like, this is ridiculous, right? And so, obviously customer expectations have changed over time. You know what? It you know, how, what's your view of kind of how you, how kind of customers, expectations of transformed and, you know, how does that influence your, kind of, strategy?

Sally: Well, customer expectation and speed of delivery of product continues to accelerate. You know, now the the one of the latest trends is, is thirty minute grocery delivery. You know, so if you place an order, it'll be at your door within thirty minutes. And to do that from a supply chain network side is it just means more locations with an optimized number of SKUs.

So it, it's a big opportunity for people and supply chain like ourselves, because you're going to have more frequent, smaller shipments from locations strategically placed near where the products will be delivered. So that has been the trend for probably since COVID. But I see that, to satisfy the needs of, of the end-consumer continuing to escalate.

Evan: Is there kind of like a bold technology bet you guys made that you feel like really paid off, right? Anything where you're like, okay, this one's a bit of a gambit, right? It's a little risky. But if we pull it off, right. Could be transformational, right? Any kind of, you know, success stories you can share there?

Sally: I think that we were very early to recognize a couple of the leading agentic AI vendors that we've had some very successful use cases that will have a positive impact on our business in terms of reducing our cost base, because we've got a tool utilizing AI to, to automate that.

So, I think that, you know, supply chain companies, service providers are typically known for being on the bleeding edge of technology. But we jumped on, that pretty quickly and are already seeing a nice impact, positive impact on the business. 

A couple of the use cases around customer service. So, checking on a customer for payment terms or scheduling an appointment or checking on the status of an order are things that we do every day through, DHL. And to have a tool to do that that's quick to implement and days and not weeks is definitely a game changer. 

Evan: Sally, how do you think about kind of developing the workforce in the age of AI? Right. There's roles are going to be changing, right? There's new skills we have to build kind of collectively as an industry in the community.

Yeah. How do you think, kind of, like what roles I play in that. And, you know, how do you, kind of, keep your, your team up to date with all the latest technologies and make sure they're being as effective as possible? 

Sally: Well, I think the nature of IT, and I look over my past 30 years and how the platforms, applications, infrastructure have changed and become more complex over time, it creates more areas of specialization.

But I think that IT people are very good at it, adapting and conforming to what interests them. And also what they see of value. I think that in terms of the overall workforce, our business partners, it's hard for them to understand the latest and to work with them on defining use cases and taking them through an implementation, I think, is the best way for them to understand the potential and light the fire in them to think of more ways to utilize the technology, because obviously they're closer to the day-to-day operations than we are in IT but sometimes they have to see it to understand it.

Evan: That makes sense. One thing I, one thing I noticed, which I found is that, it's it is hard to know how to inspire people, inspire the team, but what I've always found is, like, helpful for me personally is talking about why am I inspired.

I found what I hear, I hear other people talk about their inspirations, even it resonates for me. So, kind of, elicits the same feelings. So I have to ask, like, you're in a, you got a tough job, right? You got a big team, a complex operation. I'm sure you miss out on a lot of, a lot of hours of sleep you probably rather get, but, like, you know, you keep doing it for a long time. Like what? What inspires you? What motivates you? What kind of, like, keeps you, you know, in the game, trying to, you know, help, help your customers and help your team. 

Sally: I really like the challenge. Challenge. And no two days are the same, and I like the people and the work.

So I talk to people all across the globe all day long. I enjoy, you know, collaborating, understanding their perspectives and helping them be successful. And at the end of the day, it's all about how happy our customer is and how we're providing a service for them. But I really enjoy the challenge of making the customer happy. And how, how do we as a company leverage our resources and available technology to, to do that?

But I, I love a challenge. I love a, you know, I'm very competitive by nature and this is a fun time to be in it and supply chain. And, I love to watch the impact of, of a deployment on our sites.

Evan: Appreciate you sharing, this at the end of the episodes, we'd like to do a bit of a lightning round to try to get some, kind of…

Sally: Sure 

Evan: Short order, kind of, takes. And so basically, I would ask you, like, 4 or 5 questions that are really hard to answer, you know, briefly, but looking for like the, the one tweet response, please, you know, please forgive me if these are a little challenging. 

So, maybe just start. How do you think companies should measure the success of the CIO in kind of the era of AI today?

Sally: I think especially for AI, the ability to grow top-line revenue and not have overhead costs grow at the same rate. 

Evan: What is the one piece of advice that you wish someone told you when you first became a global CIO?

Sally: That things will take longer at the global level than they do regional. One would think that would be obvious, but when you're going through it, it's patience isn't necessarily my strong suit.

It does take significantly longer to impact change at a global level, rather than regional. 

Evan: Maybe switching more to the personal side, is the book you've read in the past that's had a big impact on you and love to hear why?

Sally: I read for escape, so I read spy novels. I like a lot of this series, and there's a series called The Gray Man. Those are very good. There are. There was one, the Mitch Rapp series are good as well, and there are several others. 

Evan: What is it, upcoming technology or product, that you're just most personally excited about? 

Sally: I love the whole Waymo car, autonomous driving. I think that is a technology I'm fascinated about, and it really, eliminates the fights between my husband and I, about how one is driving. So that is the technology I am watching closely. 

Evan: Well, Sally, I really appreciate you taking time to join us today. Thank you so much. Looking forward to chatting again soon?

Sally: Thank you, Evan. I enjoyed it.

Evan: That was Sally Miller, Global Chief Information Officer at DHL Supply Chain.

Saam: Thanks for listening to Enterprise AI Innovators. I'm Saam Motamedi, a general partner at Greylock Partners.

Evan: And I'm Evan Reiser, the founder and CEO of Abnormal AI.

Please be sure to subscribe so you never miss an episode. Learn more about enterprise AI Transformation at Enterprise Software dot blog. 

Saam: This show is produced by Josh Meyer. See you next time.