WAREHOUSE VISIONARIES
Building the Warehouse of the Future, Today with Marc Gyöngyösi of OneTrack

Evan: Hello everyone. This is Evan from OneTrack and you are listening to another episode of Warehouse Visionaries, where I sit down with the leaders shaping the future of warehousing logistics to talk about how they're doing it. So the rest of us can too. Today, I'm talking to someone incredibly smart, incredibly special.

It's Marc Gyöngyösi, the founder and CEO of OneTrack. I'm really excited for today's conversation. I appreciate you joining me today, Marc.

 

Marc: Absolutely. Great to be on the podcast.

 

Evan: Awesome. So before we dive in too far, give me the background on Marc. You know, like where'd you come from? What brought you to today? Just give me some of that background for everyone.

 

Marc: Absolutely. So the story, or really the story of myself and OneTrack starts back in about 2017. So I just finished my undergrad degree at Northwestern here in Chicago. And while I was at Northwestern, I was working on all kinds of different industrial robotics projects together with BMW Group out of Munich, which is coincidentally also where I'm from.

 

So I was born in Munich, grew up in Salzburg in Austria and then came to the US to study here, um, computer science. And so when I was working on the robotics side of BMW, we were very focused on optimizing manufacturing processes. So think about a big car factory, you're making cars, you're making several cars every minute.

 

And so every second of downtime is really important and really costly. And what we were trying to optimize for were certain robotics processes around attaching parts to windshields and car doors. And every once in a while, a robot would sit idle. And so we were looking at the data and trying to understand why is this robot not working for 40 minutes, 50 minutes at a time.

 

And well, it turned out that they just had misplaced a pallet of that specific part that they needed to assemble the car in their factory and they would literally send people with binoculars into the high bay racking, uh, on the part storage in their warehouse to find that particular pallet, that particular LPN.

 

And in my mind, there was like, there must be a better way to do this. You're investing in subsecond optimization of manufacturing processes. And at the same time, you're losing 40 minutes, 50 minutes as a millions of dollars in cost because your warehouse screwed up and how is it that the warehouse and logistics world is so different than what you have on the manufacturing side.

 

So that that's how I got into the whole world of warehousing and logistics. And a couple of different ideas and approaches at the beginning to, to try and address the problem of just missing inventory, but then over time that very quickly evolved into well, missing inventory or misplaced inventory is one part of the problem. But the real problem is you don't really know what happens inside your warehouse from beginning to end of a process, and that's what we then set out to do with OneTrack. And so in 2019, we had the first commercial deployment of our OneTrack solution. And today we're in over a hundred warehouses with thousands of sensors all across North America and Canada, tracking really everything that happens in some of the biggest operations on the planet.

​

Evan: No, that's very cool. I think there's a lot of things I want to dig in there, but you talked a lot about the problem and problem solving. And I'm really curious for your background in computer science, robotics and the engineering mindset that you bring, how does that influence the way that you approach solving problems?

And how do you think that could be translated to folks today who are trying to solve problems within a warehouse environment.

​

Marc: Well, I learned some very important lessons around that early on, and that's, as an engineer at first, I was always very fascinated by the technical side of a solution. What is the most beautiful technical solution we can engineer from an engineering perspective?

​

And one of the results of that was I was trying to solve the misplaced inventory problem with drones, like spending years trying to build the perfect drone that could fly autonomously in an aisle and take pictures of everything. But then it turns out there isn't that the label might not be visible in every single pallet.

​

So you're only able to solve a subset of the problems. You're spending all this time on the wrong part of it, which is a technical side. So the insight here was really, well, you need to understand the actual problem you're solving for the customer. And you need to walk the fine line of seeing what is possible with technology that you have, that nobody has done before, but also what's realistic to do in the real world that can actually solve a valuable problem in a business process today.

​

Over time, our thinking really has evolved. We have this mantra of think simple. We don't want to come up with the most complex engineering solution. We want to come up with the most robust and most scalable engineering solution to problems that cost a lot of money to our customers.

​

Evan: And you talked about the binoculars, right? You saw people hunting for, for specific pieces and parts. But I think there's a chasm there between you saw a problem to let me go start a company. So what was that lightning in a bottle aha moment for you? Or was there just a specific area in the, in the market that you were like, I can really address this differently than anyone else.

​

What was that moment for you when you said. You know, I'm going to leave BMW and I'm going to go start my company and go solve this problem for everyone.

​

Marc: Well, it turns out when you solve valuable problems for companies, they want to pay you money and in order to get paid money by a company, you need to start a company.

​

So that's why I started, uh, the company at the time I was, I was trying to figure out a way that we could do this for not just one company. But a couple of different ones that were willing to have me on site at the time of flying drones and then testing cameras and different sensors attached to their forklifts and their warehouses to try and figure out how do we get a better sense of what is really happening in between those barcode scans, where at the time, all they knew was at, you know, 12:25, Marc pick up, picked up a pallet this location, and then at 12:35, Marc dropped off the pallet somewhere else, but no idea what actually happened in between.

​

So it was all about just trying to find a way to work with these companies and help them, help them be able to pay us. That's what we started a company.

​

Evan: Fair enough. No, I think that the visibility I think is, is super important and something I hear a lot, just talking to folks on the show. There's so many square feet in a warehouse, right?

​

If you, if you have people on a constant roam, stuff is getting missed. So, what is actually happening because you can't really solve a problem that you can't physically see. So no, I think that's, that's super important, but you also talk through drones.

​

Walk me through the early days of OneTrack, right? What was that initial vision, and how has that evolved in those different step functions along the way to get to where we are now?

​

Marc: Yeah, absolutely. So, early days, as I said earlier, was about the drones trying to find the misplaced pallets in the warehouse.

​

And one of the challenges that we ran into was in order to test or evaluate the performance of one of these drones, or at the time I called them flying robots, somebody had to be on site, which meant that I would just always go on site, sit there with a controller, make sure in case the autonomous drone didn't do what it was supposed to do, I could take over and then land it again.

​

And that obviously was not very scalable, right? The amount of data I could collect was only, um, was limited by how much time I could spend on site with a customer flying this thing. So at some point I started to zip tie drones to forklifts. And, uh, and then wire them into the power source on the forklift to keep them powered.

​

And all of a sudden, I had a way to test tens of drones without ever having to be on site myself, because they were just on and running and capturing data. Not flying, but at least capturing the data. And then one day we got a call from one of those early customers and they asked, well, that forklift crashed into a wall and my initial response was what do you mean it crashed into a wall?

​

It's a forklift in a warehouse that never happens, right? Like no, no, it crashed into a wall and we don't have any footage of it. And like what do you mean you have no footage Yeah We don't have security cameras in the building or the ones that we had did not see it exactly We want to know what happened on this forklift.

​

Can you get us some footage from these cameras that you have on the drone? And that, that's what really opened our eyes at the time too. There's a much bigger issue or problem at play here than just the misplaced pallet in the rack. It's about seeing what is happening in your warehouse, where currently, if you're running a warehouse, you just don't have that visibility.

​

It's, it's crazy to think about, but there are, there are millions and millions of square feet of warehouse space for these logistics companies that are just unmonitored most of the time. And the only thing they know is, well, Marc picked up a pallet at 9:25 and dropped it off at 9:35, but those 10 minutes in between, no idea what actually Marc was actually up to during that time.

​

So that, that was that aha moment for us in 2018, 2019, when, uh, we realized there's a much bigger opportunity for us here in terms of bringing visibility. Not just through trying to find something in the warehouse, but actually tracking what is happening inside the four walls of a warehouse.

​

Evan: That all comes back to technology alone can't solve your problems, right? It's that human, the loop piece where you need the people to actually buy in to the processes. Because if everyone's followed the, the SOPs in a building, the building's going to run itself. You're gonna have no issues if everyone does all the things they're supposed to do it's it's when people don't, that those processes break down.

​

So technology is really supposed to be an enabler to help you find the reasons why those processes are breaking down. I think that's a really important piece of what we're doing here at OneTrack of actually uncovering the, the human behavior, those leading indicators of why something happens today that then causes issues hours, days, weeks, even months from now, in case of like shipments going out with problems.

​

Marc: So here at OneTrack, right, our customers are either 3PLs or they're shippers. They're the manufacturer of the product. And, and, and if you're a 3PL, your core business is people, process and technology. That's what you do to deliver, deliver service to your customer. And the level of service that you're able to offer to your customer is exclusively a function of those three things.

​

And you cannot just focus on technology. If you don't focus on people and process, you can solve your problems just by focusing on people because the complexity of modern operations. And that's why companies like us exist today, have evolved quite a bit from just 20 years ago. Expectations to logistics companies have changed a ton.

For example, also as a result of how Amazon has changed how warehouses are run nowadays, and the expectations of consumers on how product needs to get shipped to and from them, right? So there, there are a lot of different factors at play, but you can't look at any single problem or challenge in a silo.

​

You always have to look at it in the context of people, process, and technology. And you need to bring something that actually ties those three pieces together so you can drive service quality, which ultimately drives your margin as a 3PL and, um, also as a shipper.

​

Evan: So let's shift gears. OneTrack calls itself a Warehouse Operating System, which I think can mean a lot of different things to a lot of different people.

​

So. For you, break down what is a Warehouse Operating System and then what is the vision for the warehouse of the future operating in this kind of using technology like this, people like that, process like that. Has it all come together into this Warehouse Operating System vision?

​

Marc: It's interesting that you say the warehouse of the future, because I think when you walk into a warehouse nowadays, today, it doesn't even feel like it's today.

​

It doesn't feel like it's 2024 when you walk into a distribution center, into a warehouse, even just today. So when we think about the future, I think the future is defined by trying to catch up with what is possible in technology and automation. From just a current state of warehousing side of things.

​

So when we think about Warehouse Operating Systems, we think about creating a single source solution for our customers that allows them to manage how people and processes function within their operations. And use technology to monitor how those people and those processes are functioning. So if you are a 3PL, you can understand across my 50, 100, 150, or even if it's just five sites, how are my supervisors, how are my managers operating within this building?

​

What are the exceptions that are occurring and how are they responding to those? If you're a manufacturer of product and you look at your supply chain and you look at many different logistics partners across your supply chain, you're interested in understanding, are my partners following our process expectations?

​

Are they meeting our contractual goals around the productivity around, but also around things like safety. And then when it comes to topics like safety. Are they, are they making sure that the employees who are working in buildings to move my product as the manufacturer, are they following the, the right expectations around coaching and development of employees?

​

Are employees being effective and safe at the same time to perform their duties and move our product across our supply chain? And then when you think about it from a product side itself, well, what does our product actually look like before it gets to our end customer? But if you're selling, thousands of pallets of cereal to a, to a big customer, you want to understand what do those pallets actually look like before they get shipped out to that end customer? Not just at that point, but what does the product look like when it comes off the production line, when it gets shipped to that distribution center, maybe it goes through a cross-dock environment and then it goes somewhere else.

​

And if you could have visual evidence of what actually your product look like and your orders looked like as they move through your supply chain, there's incredible value that I can uncover and the ability to drive a margin in your processes.

 

Evan: And I think some people might argue, well, that sounds dangerously close to a WMS or, you know, just your typical warehouse management system.

​

And some vendors might claim to do a lot of those things. So how do you differentiate OneTrack from a WMS? I know there's some tie together there, but where do you see the places for each technology? And especially as things move for the next five years, how do you see that playing out?

​

Marc: The WMS is very good at keeping track of transactions.

​

So somebody scanned something from one place to another. Historically WMS have been very challenged with a lot of different configurations that are required to make the system work in different places. Every single customer, the moment you bring up the idea of WMS integration, you can, you can sort of see like everyone looks out and they get worried about how expensive is this going to be?

​

How long is it going to take? And the reason for that is because a lot of the WMS software out there, just, it goes back to the comment. It doesn't feel like it's 2024. It's software that's not built in a way, uh, that we would expect to see software built today. So it's not very easy to configure and adjust it for the different nuances and update it as well as your business and as your operations change.

​

So we think about the WMS as a source of context for us today, meaning that from the WMS, we know what is the activity that is being completed by the employee. But then the big difference between that and OneTrack is that OneTrack is using visual data, visual intelligence to understand Not just the timestamp of a, of when something was scanned, but what was actually happening, happening on and around that particular lift, that particular employee at this time, how did they move through the building?

​

Did they take the fastest, most efficient route? How many stops did they make? How did they pick up the product? How did they drop off the product? Did they bump into any racks along the way? And then more important and very importantly too, when that palette gets put on a, on a trailer on a truck, what did the palette look like?

​

So fundamentally WMS cannot support that kind of data today. And we've built a technology that allows our system to capture the visual data and then learn from that visual data to identify things in there that is actually relevant to pay attention to. So the WMS today is, it's a record of truth of what happened, but OneTrack is a record of truth of why things happened.

​

And the why is often much more important than the what, because if you don't know why something happened, you can't fix the problem. You can't fix the problem you can't see.

​

Evan: And I think that goes back to, you don't just want something to be a rearview mirror, right, of what happened, because rearview mirrors are not very good at looking forward and helping you plan what to do next.

​

So, when I think about someone preparing for peak season, you can only do so much with historical data. You also need to be able to see, okay, what's coming next? What are the trends? How is my building operating in real time? And ideally have that all in one place, so you can see. and hold people accountable to your process and performance and make adjustments on the fly.

​

I think that is kind of the future of what warehousing is moving to.

​

Marc: Everything in this world is moving towards artificial intelligence as driving decisions, processes, people. The problem in logistics is that you fundamentally cannot build good AI systems on top of the data that exists in the existing systems there today.

​

So if you were to try and build any kind of AI application on top of your WMS, but you can't even answer how many pallets exactly that we shipped yesterday. You are not going to get good insights from that AI algorithm, from that AI system. So fundamentally, and that's the core premise behind OneTrack, what we are starting with is we're building technology that allows our customers to capture accurate data in a scalable way across their operations in and outside the warehouse.

Understanding how is the product moving? How are the people moving? And with those data points and with the visual intelligence that we're delivering, now you can actually leverage all those new AI innovations and improvements that are out there to, in a, in a accurate way. Accuracy is the most important thing.

​

It's very easy to create demos that are really convincing in a very controlled environment. But if there's one thing true about logistics, it's not a controlled environment. Sure, it's a warehouse with pallets moving in and out, but every single warehouse is different, even within a single customer or a single network of a customer.

​

Evan: So what do you think is the biggest challenge that leaders face or, you know, the general manager at a warehouse? What is the biggest challenge that you see them facing when it comes to bringing technology like OneTrack into the industry, into their, their site, across a network? What are some of those roadblocks that you're hearing from customers that you see that are keeping them from making that leap, aside from obviously finances and some of those business metrics?

 

Marc: The return of great technology in this space is so significant compared to the cost of it, that that is rarely ever the reason why a system like OneTrack is not successful. The real, the real thing there, the real bottlenecks and the real block roadblocks are, comes back to the three pieces, people, process and technology, right?

​

So you, you got to have the right people in place that are willing to change how they operate their business today and look at things through a different lens. You have to have that there. You have to have the right culture in place as well, where your employees are able to look at new data that they never had access to before, not in the context of how do I now blame someone else, but in the context of how do I now use this insight, this visibility, to work together and find a better way to complete a job or a task, right? That's really key when you bring in new data. It's not about finding who's, who's guilty and whose fault it is. It's about identifying how do we work together to make this, this operation work smoother, safer, and more efficiently process.

​

So you can give somebody, imagine, you know, it's 1995 and you're, you're down, you're using Microsoft Excel for the first time. You've never used a spreadsheet before. Well, that's not that easy, right? If you don't have the right process in place to understand, where do I use this new tool that was given to me to be effective at my business?

​

You won't be able to get to how some, how you are using Excel today and optimizing or understanding data and running analysis, running queries, looking through giant sets of data. Those are, those are things you need to learn. You need to build process around. And that's the same thing with new technology, like OneTrack.

​

When you bring that into your operations, you need to make sure that your people are trained on using it and that your processes are updated. So you now clearly state that when you think about. Safety observations. You are now using this new technology in the context of safety observations. It's not just something that every once in a while we take a peek at it and don't really use it for anything else.

​

You need to be willing to make the commitment to update your processes to reflect on the abilities that are now available to you and your team to use this new technology. And then the last part is technology itself. I talked about drones. Drones are really exciting. They're also just that, they're really exciting, but is it really the best way to solve this problem?

​

Probably not. We learned that the hard way and with any technology that you bring into, into a business, the important pieces is the technology. Does it fit into how we operate as a business, right? Is it actually at scale more efficient and better than what we're doing today. If it's not, then you're just doing it to take three steps forward, five steps back.

​

You want to find technology that can support you along every step forward, and it doesn't force you to take steps back. You don't want to have to just reallocate resources from people that drive forklifts to now technicians that need to maintain your autonomous forklifts. Guess which one is harder to find, right?

​

So the, the important piece here is the technology needs to be robust enough that you can use it at scale, and it has to be robust enough that it delivers the insights you expect to be valuable at scale as well.

​

Evan: And you brought up something interesting there in the technology piece. I think everyone is talking about the rise of robotics and AGVs and just automation within warehouses.

​

What's your stance on that today and where do you see that changing in the future? And then I have a follow up to that based on I think what I think you're going to say, but I'm curious to just hear your opinion on the rise of robotics and the industry craze that I think is the other half of the AI coin in the warehousing industry.

​

Marc: Robotics is super exciting. It's also been around for a really long time, longer than I've been alive. There have been autonomous robots driving around in factories and they're older than I am. And so the, the reality I think of robotics, obviously there's a lot of new technology now, again, in terms of the ability to better understand the environment, be more flexible and deployed in different, different places.

​

But you need to look at what's actually the, what is the true ROI on the equipment? That I'm, that I'm getting. So I think there is, there is a lot of really interesting, uh, there's a lot of really interesting technology out there around the fully automated warehouses where you have full control of the, you basically just have a Product show up and it goes into some giant system.

​

It's fully automated. And at the end, you get a bunch of items back on a pallet like it's shipped out. That makes a lot of sense. When you have the ability to make that long term commitment, you know, that for the next 10, 20, 30 years, you are going to be moving this kind of product through these DCs and your, your packaging is not going to change significantly.

And you're able to support the variants that you expect to see in your business. If you're a logistics company today and you have a three year or five year contract of a customer, you're going to have to look at more lightweight solutions because you can't make that giant capital commitment of spending 50 to a hundred million dollars in a fully automated warehouse.

So there are certainly technologies out there that could be suitable for those, for those environments. But I think the biggest issue there is always the scalability of it and the reliability at scale. Is that shrink wrap on the pallet going to stop the robot from driving down the aisle? And then when the robots stop somewhere in a million square foot building, you have to first find it.

​

And then first you need to figure out that it happened, that it got stuck. Then you need to find it. You need to send a technician there to fix it. You're kind of losing, it's kind of a moot point now around the robot, because now you have to have people support the robot to do this thing that they're supposed to do autonomously.

​

So that that's where I think it's, we're still learning a lot in this industry about around the robots. And I'm excited to see more and more of those robots and our customers warehouses too. But the reality is there's so much out there that you can solve right this second without having to wait for giant advances in perception and control for robotic systems.

And, and that's where, that's where you can actually make money as a business. That's where you can bring value as a business. If you're a logistics company, you can offer your customers. Here's a picture of every time somebody touched your palette as a service. That's not something you need to bet on some giant advance in robotics to happen. That's something you can do today. And that's what I would be focusing my time on. If I was on the, on the logistics side of it.

​

Evan: You actually answered my question there. I was going to ask, um, what would you say to someone who is trying to choose between investing more in robotics and saying that, well, I think our warehouse is going to be X percent more automated in the next five years versus Someone who needs to weigh out, why should I invest in that kind of human in the loop coaching that, that OneTrack brings and how do you balance those two?

​

And I think you kind of hit the nail on the head there of, you need to invest in solving problems today. And I think in the 3PL world, the business model doesn't always make sense to do this massive fleet of, of AGVs and automated vehicles in warehouses because of the, the payback and how you have to deal with the contracts with customers, then relaying out a warehouse for specific customers.

​

It's just the unit economics don't always work out in the end there.

​

Marc: There's this fundamental assumption that the warehouse of the future, which let's first focus on what the warehouse of today looks like. But if we talk about the warehouse of the future, it has to be fully autonomous and automated. I don't think that's true.

​

I think if you can increase the productivity of people by 30, 40, 50%, the business case for robots gets that much more complicated. Because all of a sudden, your whole model and your ROI assumptions are different. And guess what? When people are 30, 40, 50 percent more efficient, they're probably also happier.

​

Because the work works better. And when you think about what is happening inside a warehouse day in, day out, there is so much low hanging fruit, you know, one bite at a time. How do you eat a big sandwich? Well, you don't, you know, try to eat it all at one. You take one bite at a time. And I think the same is true in a warehouse.

​

You look at your lofty goal of 30, 40%. Well, let's start with the simple things. Let's start with, is everything zoned correctly? How much time are people waiting for things? Are things stacked correctly when you pick them up or do they have to get off the equipment and like reshrink wrap things? I think there is so much time lost on all these small things that if only you could see and fix You're going to take, get there much more reliably than trying to make this giant bet on something that isn't really proven out yet out there.

​

Evan: And I always like to throw a curveball here to to round things out. So when you think back from the early days to now, what is something or maybe it's a few things that you've learned since working in the warehousing world that maybe you didn't expect or caught you by surprise, or it was just kind of a just just a curveball that got thrown your way that you've had to adapt to.

​

Marc: Pallets fall on forklifts often, and I really did not expect this. And the first times when it happened, it really caught us off guard because we used to have cameras pointing up, uh, sticking out from the overhead guard of the, of the forklifts. And every time a pallet fell onto those overhead guards, the whole thing broke.

​

And we had to go back on site, replace the camera, and then we tried to build cages around it and protective covers and all kinds of things until ultimately now our sensors are completely broken. super small and they just fit nicely underneath the old guard. There's no damages from it anymore. But that was really one of those things where I did not expect it.

And I think a lot of our customers, when they see those kinds of events for the first time, they also don't expect it. And they especially don't expect the severity and extreme risks that comes with an event like that one. I mean, think about it. There's a giant palette of product. It could be It could be sulfuric acid if you're in a hazardous material warehouse. It could be peanut butter if you're in a food warehouse. It could be tires. It could be engine parts. It could be clothing. It could be underwear. It could be toilet paper. Whatever it is. Cigarettes. Whatever it is. But that palette just falls down onto the overhead guard of the forklift. You have this giant spill of product all around you.

​

And often you don't really know what happened and why it happened. And so now what we have is, uh, when we do get sensors back from the field, let's say something falls onto it. We have a special place here at our office for those devices, because sometimes, you know, the, the food product that might fall onto a sensor like that, uh, not the greatest smell when it comes back, uh, days or weeks later here to the office.

​

And we want to make sure our software engineers sitting at the at the desk don't get distracted by the smell of uh, some food product on on a sensor that came back from the field because some pallets fall fell on a forklift. But yeah, that's that's probably one of those things where I would have never expected it and until you see it It's hard to believe.

​

Evan: You can't solve problems you can't see so I guess that that sums that up.

​

Marc: Absolutely. You have to see the things you want to solve.

​

Evan: Well, Marc, thank you for today's conversation. I think I learned a lot. I'm sure other people will too. I appreciate you coming on today's episode.

​

Marc: Absolutely. Thanks for having me.

​

Evan: And everyone who's listening, thank you for tuning in to today's episode of warehouse visionaries.

Remember to check us out at OneTrack.AI/warehouse-visionaries, or just follow us on YouTube or LinkedIn. And of course, don't forget to go find Marc on LinkedIn, give him a follow and engage with him. We'll see you next time.

​

This was an inspiring conversation with Marc full of actionable ways to improve your warehouse operation. Here are my top three takeaways.

​

Number one, think simple. You need to find scalable, cost effective ways to solve problems. It's not always about the flashy demo in a controlled environment. It's about real value, despite the imperfection that everyone knows comes along with the warehouse world.

​

Number two, you can't solve problems that you can't see. There are so many things that happen every day in warehouses, but until you see it You can't solve for it. You have to invest in technology that brings real visibility beyond just data in a spreadsheet.

​

And number three, technology alone won't solve your problems.

​

You have to focus on people, processes, and technology equally. Without one, they all fall apart. Whether you're a 3PL, a shipper, or a manufacturer, your business relies on all three to be successful.

​

For more ways to take your operation to the next level, join OneTrack's newsletter to get emails twice a month full of real stories and advice to solve your biggest challenges.

​

And subscribe on YouTube at OneTrack AI to never miss an episode of Warehouse Visionaries.