Take the Blindfold Off Your Warehouse Operation
In this episode, Kevin Coleman, CEO of CJ Logistics America, shares how his team has approached AI across a $1 billion+ logistics operation.
Rather than hype, Kevin focuses on the realities that most traditional industries face: inconsistent data, real security concerns, and the risk of chasing new tools without a plan for scaling. He explains how CJ Logistics built its foundation, adapted for broader operational intelligence, and balanced innovation with compliance while managing a workforce of more than 4,000.
In this episode:Â
- Why quality of data matters more than the sophistication of the model
- How the company evaluates new technologies before deciding what to scale
- Balancing team innovation with safety and operational constraints
- Expanding to integrated operational intelligence
For leaders navigating AI adoption, Kevin’s perspective is a practical guide. It shows what works in a traditional industry, what doesn’t, and why disciplined execution matters more than noise.
About CJ Logistics America
CJ Logistics America is a North American-based integrated supply chain service organization with operations in the United States, Mexico, and Canada. The company offers warehousing, transportation, and freight forwarding services for all temperature classes (ambient, temperature-controlled, and frozen). CJ Logistics’ customer-centric philosophy ensures that all solutions are tailored to meet clients' specific goals and objectives.
‍
Transcript
Marc: Kevin, thank you so much for joining me here today.
Kevin: Thanks, Marc.
Marc: I love to start out conversations with the question of what got you into logistics, because one time someone told me, once you're in logistics, you never leave. What was that moment for you when you ended up in logistics?
Kevin: I did not go to school for logistics, but I'm a little older than most, so they didn't really have supply chain programs back then. But a gentleman recommended me for a consulting role in a firm—kind of a boutique industrial engineering firm—and that launched my career into logistics. So I spent my first eight years in consulting doing strategy, operations, and technology, and that really shaped my foundation of logistics and supply chain.
Marc: What do you do today?
Kevin: Yeah, so just a quick journey—I joined a domestic 3PL. I was in the account management group consulting, again doing a little bit of solution design work, took over sales. And then we were acquired by CJ Logistics. And so I've been the CEO of CJ for the last several years.
Marc: And can you tell us a bit more about CJ Logistics and what CJ Logistics does and how that role also has evolved over the years?
Kevin: Yeah, CJ Logistics—we're part of a large holding company that has many divisions where we roll up to the logistics division. We have a little over 9 billion in revenue, broken up by our APAC division, our IFS, which is our freight forwarding, then our Americas division, and we're about $1.2 billion here in the Americas.
Marc: Can you give a sense of the kinds of customers that you serve, the kinds of products that you offer those customers and the services?
Kevin: Yeah, so we focus mainly in the consumer goods area, but we also have the automotive space as well. We're in the US and Mexico and Canada, and we focus on three core segments: our warehouse management segment, our transportation management, and then our freight forwarding business. But again, we really see ourselves, as we've evolved our company, really more into an integrated solution provider. So we have different products along those continuums and products that we continue to develop to meet market demands.
Marc: Can you tell us a bit more about how those relationships have evolved over time and what role technology really plays in building the relationships with the customers as well, and maybe their expectations and how data has evolved over time?
Kevin: Yeah, I mean, technology's always been critical in supply chain. Other spaces have evolved quicker than supply chain. Supply chain's a little bit of a lagging industry when you think about just retail and some other things and how tech has taken over in some of those categories.
But to get back to your question on customers and maybe 3PL a little bit—you have to have tech to play, so to speak. And not only have technology, but you have to have an essence of what is your roadmap and what does that roadmap look like for your different core products.
And I think what's been the biggest change is we see a lot of really unique and interesting technologies coming into the supply chain space. And then how do you select those and how do you make those part of your larger foundational portfolio? And I think that will continue. And I think that's good for the supply chain, because I think smaller companies can be a little bit more nimble in areas that can make a great impact to our business, but also our customers.
Marc: And you mentioned there that in the world of logistics, it's been a bit lagging on a technology side. When I first spent a good amount of time in warehouses, I was always surprised where on the one hand you have all these technology vendors and robotics and automation projects out there where people are talking about the future of logistics. You know, everything will be automated at some point in the future, but so often I found that if you just focus on right now, today, let's make sure that we get the operating systems of a logistics provider into what is already possible out there on the market, right this very second. So like, not just thinking about the warehouse of the future, but just let's make sure the warehouse of today matches the capabilities of the technology that exists today and bring that into the real world. So do you have some examples of technology that you implemented from that side?
Kevin: We look at it two ways. I think in my seat, you're right—I think you have to look at technology of today. I think you also have to keep your eye on the future. So we do have, where it makes sense with customers, maybe with long-term contracts, or we have—I don't know if I would call it warehouse of the future, but fully pretty automated facilities.
And so, but there's a lot of other facilities that don't have that commercial arrangement or the length of agreement, which means you need to find other ways to engage technology into those operations. So, quite honestly, that's what we found with your product, One Track, a while ago. I really don't even—can't remember when we started, but it was probably pretty early. Probably one of your first customers. I mean, I'm not sure if we're the first, but we're one of your first.
And we saw that as a technology that could fit the models of a lot of what our customers look for—three to five year agreements. We thought it was very flexible for us in our ability to maybe move that from site to site or expand up or down, based maybe on peak and non-peak season. So we felt it was a very flexible technology, which is critical in a 3PL when you don't have a very, very long-term contract. So I think that fit really well to our strategy, and I think we've had a great partnership over many years.Â
Marc: CJ Logistics has been a phenomenal partner for us as well. See how we can take the technology, which originally started with a fairly narrow focus on safety, and then now over the years has evolved into a much broader look at what happens inside the four walls of a warehouse and soon also outside the four walls of warehouses. But the key thing really is the data capture side of it, right? As I think about the role of One Track, our role is to capture the data and give you access to capabilities that drive intelligence with that data to your customers. But then an interesting perspective there is, as you think about the role of a 3PL and working with your clients, you're almost like the—if the economy is a bloodstream, you're measuring all the blood that's moving through all the different veins of the economy in a way, right? Like you can see what happens in the real world. So how do you think about that as it relates to relationships with your clients?
Kevin: Yeah, let me address—in the industrial work, what keeps me up at night is safety always. So you know, we have about 3,500 employees and it's critical that the way they come to work, we send them home and they can enjoy time with their family.
So if we go back to why we initially went with One Track, it was to improve our already great safety record. And I'm very proud—with your product and tech, we've been able to improve that. We're running at 1.60, which is phenomenal in the space. And we were good before.
To answer your second question, you kind of alluded to—and I have a saying—I really think your product helped us get there originally with safety. But supply chain or logistics has a ton of data. It's limited on their information and it's harder to get to the intelligence piece. Where I think AI plays a huge role is that bridge to getting intelligence at a much quicker pace and a much greater scale. So I really do feel, in that realm, I see a huge opportunity for AI across all of supply chain activities.
Maybe the third point, and maybe it's not exactly answering your question, but I think it is with customers and how we have limited resources. And then so when we look at business, we select products—are there other use cases out there that we can evolve with our partners? And I think that's also been a great success, not only for us but for our customers. So I think that's been a good partnership—your openness to engage with us beyond safety and to other use cases and really then take that value back to our customers.
Marc: Yeah, everyone is always excited about ChatGPT and the LLMs out there, and it's moving at a pace that is faster than really anything we've ever seen. I talk to a lot of people who say they would work on technology during the 2000s when the internet came around, and they say the internet was nothing compared to what's happening here right now with large language models and the ability to just prompt AI to do work for you. Analyze data. But the thing that so many people miss is if you don't have the right data, if you don't have access to what actually happens, you can't really do much. Where the role of the 3PL is so unique because you're right there. You see it all. You see all the product that moves where your clients—you still help them.
Kevin: You still have one of the basic problems and you know—garbage in, garbage out too. It really drives now more than ever. So just having a great data structure, and that really starts with our customers too—we're making sure we understand the data streams and our technology's only going to be good as our foundational data and our structures.
Marc: And then can you talk a bit about CJ logistics as a whole and the different products and the way you're thinking about the future and how those are evolving from your core to the other areas of the business?
Kevin: Yeah, first and foremost, we always start with our people—very important to us. And how do you drive a business? And then how do we create customer value? And so at the highest level, we want to do a couple things. One is we want to lead growth and change for our customers. It's coming at a faster pace than ever. We want to transform business process and maybe processes that have been in the supply chain—I think your product helps us do that. If you talk about process being safety and how we've transformed and how we use your tool to interact with our customers. We want to reduce costs for our customers. Total cost—that's what supply chain's all about. How do you reduce and improve service to our customers? And I think if I talk about some of the use cases with your products, I think we're starting to get into that service area as well, that quality, using vision in a couple different ways.
So from a product standpoint, we look at our business—we have three major categories: warehouse management, transportation management, freight forwarding, and as I mentioned earlier, from that we look at our core, adjacent, and transformational strategies. So for each of those products, we have set unique strategies for those markets that either align with commercial demand or customer demand or future demand. So that's what we're really looking for and that's how we set our strategy.
And then I think that's how I see our partnership with One Track evolving, quite honestly. I think our core was safety. And then as we look at other use cases, how do we look at the adjacencies that you can provide value to our supply chain? And how are some of the transformational things really down the road? Yeah, maybe we don't implement next year or year after, but we got to start thinking about those. And how does One Track play into that? I think is really critical. Just like our product strategy.
Marc: And then some of those use cases we've tackled together after the safety piece, especially productivity has been a big focus, right? So as you think about the different levers you have to optimize for your customer, obviously you can optimize at the network level. You, yourself as a 3PL can optimize at maybe the asset level or the site level, but then ultimately when you're inside the four walls of a building, you're really trying to optimize the processes within that very building. And the interactions and the work that happens every single day between the people. So can you maybe talk a bit about what it means to have more than just a barcode scan and a timestamp to look at those optimization opportunities and the processes and what's really driving cost and efficiencies?
Kevin: Yeah, I mean, if you look at warehouse, especially in our environment or where we come from, it really starts with 70% of all expense is really travel distance. And so how do we use your technology working with our engineering group to reduce that travel? There's other technologies we use to do that as well, but that really creates efficiency.
People talk about cost savings. I like to talk about waste reduction—travel's really just waste. And so how do you continue to drive that waste out of the supply chain? So that's really where we like to start—how we optimize our buildings through engineering, using technology to really get that waste out. And then we have other ways we're able to evolve that into productivity by function to help us really address how we man the facility and how many people we might need, the different structures, which then benefits our customers.
So it's that whole—I like to call it optimization of a warehouse and really elimination of the waste—yeah, we've really taken that use case, I think, together. And if our core was our safety, I call it pretty much an adjacency to what we were working on as our core and what we've talked about in the future. And we've been doing some of that today even, and I don't know if it's transformational yet, but quality as well. So I think there's some—with all the vision and all the data we're collecting, there's opportunity.
Marc: And one thing I've noticed, especially when you think about the optimization side of things in the past, you said "garbage in, garbage out." Okay, so let's say you start with good data. So you have good data. Now you are trying to make sense of that data and understand what's actually going on in this warehouse. And where are the optimization opportunities that would typically require resources to spend time, dig through the reports, maybe create new reports, new charts. People would create their own spreadsheets and macros and manually import and export data. And nowadays we find ourselves in this unique position where if you set the systems up correctly, a person can just ask a natural language question and just ask a question like, "What's my gap time yesterday and who should I talk to?" And you don't have to log in anywhere. You don't have to click on any buttons. You don't have to learn how to navigate different filters and menus. The answer just comes right back to you all the way to historical context for the employee that you're supposed to talk to potentially about that or a process maybe that could be improved. So that really changes, I think, the capabilities of the people that you have working in the warehouses as well, not just on the forklift, but the managers and the supervisors and all of those. So how do you think about that and like the upskilling essentially of your teams out in the field?
Kevin: Yeah, I mean, it goes back to our earlier dialogue and conversation. I think where AI is placed, not just with your product at all—the data, the information, intelligence kind of piece, right? So it gets you the intelligence a lot quicker, right? Before you had to download it, you built some Excel and whatever you would do, and do some analysis and you get to the intelligence eventually. And the AI component really gets us there quick.
So then it's just how—to your question on upskilling talent or however you look at that—I look at it more as, how do we embed that into our leader's standard work? So how do we not throw so much at them, so much data. What you need to provide is linkage to leader standard work. So what am I going to come in to do for the day? So it's very digestible in how I go and lead my day. So it's great to have all these things—oh, I just ask a question—but when do I ask the question? Or how do I ask the question?
So I think embedding that into leader standard work is what we also do—how we use your tool, when we use it, where we go is also critical, because you need to kind of have that end-to-end view. So I kind of—I don't know if it's upskilling as much as making sure we're using tech in the right spots within our operation, and that we're giving good training and development through our leader standard work to use those tools appropriately and when to use them.
Marc: Yeah, it's like every tool in the world—if you don't use it, it doesn't help you, but if you know where and how to use it, it can really make you exponentially more effective.
Kevin: And I think that—I used to, long, long time ago, implement technology and I think one of the greatest challenges of tech is you've got this big bucket of functionality and sometimes you only get to 60%. And you leave a lot on the table. And that's not because the tech is bad or the people are bad, but I think it's the intersection which makes—it's so complex—you got, I mean, the old saying of you got people, process, technology, and I think that's where you take tech. You layer in your leader's standard work, you layer that in as it's a process, and then you lay that in with good training and development that kind of builds that.
I mean, trifecta is probably the wrong term, but just that's what you really need to focus on. I think as a business leader—it's not always the technology's fault. It's not always the people. It's just the intersection of—and how do you use that together to get the most out of everything?
Marc: I want to take a quick step back and just look at AI a bit more holistically. And the last four years, I think really everyone's mind's changed every couple of months on what's possible, and it just keeps getting wilder and wilder, the capabilities that are out there. What was your first aha moment on actually using AI, not just as a chatbot, talking to it in ChatGPT to write some email or something, but actually do something where you think, oh, in the past I'd have to go and dig into this myself or ask a person for an update. I can literally just ask it on my own schedule and get the answers right there and then?
Kevin: Yeah, I mean, AI's everywhere, right? I was probably maybe a little lagging adapter a little bit. But for me it's every day. Whether it's email or reviewing our CRM data, or to your point earlier, it's moved from me having to make a phone call or send an email or a text to get the information I need to me being able just to ask a question along multiple platforms as long as the data is there.
Yeah, is there garbage in, garbage out? It means the fundamental piece of this, right? So you got good quality data, it cuts down a lot.Â
Marc: Yeah. And if you think about that, maybe three years ago, we probably would've never imagined these things to be possible. And now if we think forward three years into the future, I'm sure you're thinking about this in many ways, also about going back to the different products that you're building out and growing. How you build those and how you go to market and how you position them in a way that might be different than you would've done in the past. Right?
Kevin: Yeah, I mean, we've really changed, even just—we try to look at a lot of trends in the market and look at that monthly, quarterly, however you want to look at it. We just adopted a new tool where instead of trying to build a report, what we load up is a new AI tool, which is all the different market data sources we use. And then to your point, we just ask the questions like, "What is the growth in the 3PL industry? If you're going to expand in the warehousing space, where would you look?"
And it starts to give you those answers and you get the full article, but it's really nice to be able to—so there's a lot of ways that we use it in the business.
Marc: Testing hypotheses is one thing that I find quite fascinating. Where you can sort of play through different scenarios and play them forward. If I did this, what do you think are three scenarios that could happen? Like, okay, let's play this one further a bit more. And it really allows you to operate the business differently as well. It challenges your assumptions.
Kevin: Mainly your bias. Based on what or where you're coming from, I think it helps you challenge that and maybe think about things in a different way. And I think that's good as you put your strategy or project plan together.
Marc: One of the things we talked about earlier was multi-client facilities. Can you maybe talk a bit more about how you think about that and why that's interesting, but also what the challenges might be that come with that?
Kevin: Yeah, I'll start with our core, adjacent, transformational. We started in dedicated warehousing and we've evolved. We're probably always into multi-client, but really from now we're making that play to grow that space. The final front, transformational is really our cold chain space, but let's focus on multi-client per your question.
What we've seen a lot in the industry is—before a lot of large CPGs, they still require dedicated space. So you still do a lot of dedicated warehousing, but even some of those larger CPG consumer goods companies are looking for that shared space environment and a little more flexibility.
So the biggest concern when you get multiple clients in the building is how do you manage the labor? How do you manage the volume?
Marc: Different WMS systems.
Kevin: Yeah. We do have some of those with multiple WMSs. We do try to encourage a singular WMS, and so yeah, a lot of different challenges. What we've done too is where I think AI plays in the future—I know it will—not only in our multi-client facilities. But if you look at our dedicated facilities and our frozen facilities, all of those require a level of shipping, receiving, and customer service. We've started a shared services program, which really unlocks some of those functions that can be managed more on a central basis.
And you still need stuff in the building. But you can really extract some of that, whether it be inventory, customer service, shipping, receiving, using technology. And so that's really been powerful for us as well.
Marc: Right. I mean, if you think about shared services plus AI to do a lot of that work where you can be in a central place optimizing tens, or if not hundreds of warehouses across the country and operations without ever having to be there, because you can have sensors in place that can tell you what's actually happening, and then with AI, that's digging through the data on its own, doing potentially thousands and thousands of hours of work at the click of a button. Now all of a sudden you have exponential capacity and capability on a team that in the past would have to maybe travel from site to site.
Kevin: What you know, that's where you're getting into the transformational category. Exactly. You have the shared services we move from, and we tell you what we're doing and you're able to jump in and probably help us there as well. So I think, yeah, you got to think about the future and we—it goes back to—I hate to be that—you talked about data and the foundational—before we could jump to that shared services. I mean, you have to get some better consistencies around processes and data.
So that's really what we've been able to do is consolidate those processes and data, and I think having that foundation in place is really critical, I think, to get the full value of an AI platform.
Marc: So as you think about—you can never get in too soon on AI, but if you were to talk to someone who doesn't have any AI yet in their business, what advice would you give them?
Kevin: I think we had a couple starts and stops. I really do think it depends on the industry, and this is probably a bigger answer than just AI. So when we look—how we evaluate—we have four centers of excellence here at CJ. One is our technology COE. And that helps us—there's a lot of noise in tech and so it helps us separate some of the noise. But one of our foundations of the COE and tech steering committee is—can, how would that technology not only be applied for this use case, but can it scale?
It's really important for us. We can't—if we just had you in one building, it doesn't work for us. Right? So you being a very large part of our network and some anomalies why you're not, but for the most part in every building allows us to scale.
And then also we don't have armies to go out and invest across. So when we look at our COE and our tech—are there other use cases we can expand on? So my advice going back to your question would be, look at the tech, whether it be AI for the use case you're trying to solve. Can you scale with that across—if you have multiple buildings or whatever that might be, and then does that provider have the opportunity to work with you on other use cases? So as you need to go back to the drawing board to get another evolution of that product, that same process. Can I scale it otherwise? My advice is it's too easy to chase the shiny object. And then that can get you into a scale problem and then also a resource problem, because you're expanding now with different product vendors to really get to what you need to from a scale perspective and scope.
Marc: Yeah, I think that's a really key piece that the LLM doesn't care if it's analyzing WMS data or your customer order or something completely different. It's just a matter of creating the context around that data and then the tools and the infrastructure for it to access it reliably and process it.
Absolutely. And I think from a One Track side, over the past six months, our perspective and our view of what we can be for our customers has changed quite a bit as well. The first, call it, eight years or nine years of building this company, it's been function by function, safety, productivity, quality, like building out those tactical pieces. But what we're realizing now is if you can become the AI platform for the customer, it's an incredibly interesting position to be in for us because we have the physical sensors that are capturing the data that are actually telling us what's going on, but then building additional layers on top of it is just a matter of bringing the right data into context.
That could be from a customer system, that could be a system that's built in house in the future. It could be some new sensor modality like we recently started to roll out temperature and humidity monitoring sensors. Like now all of a sudden all the warehouses that have those sensors will alert on floor sweating. And yeah, like one small use case there, but in the past you'd have to build that solution as a one-off thing. And now it's just, let's just make that data available to the AI agent and they can do the functionality that other stuff would've had to do as a point solution in the past. Yeah, and that's really exciting. We can usually do so much more with the same core.
Kevin: Yeah. And it's a good point. And we go back to a little bit of the other question. I probably—the gap I missed in what else advice. I probably visited 12 facilities over the last four weeks and had another one tomorrow and a few next week. And just ask the users—we can, you and I can sit and talk about the greatest technology in the world, but if it's not getting utilized, the usage isn't there, the compliance isn't there. The users usually have really good ideas of why they're not using it.
Sometimes they're bad, but mostly they're good. And then how do we work with them and make sure the adoption's there? That's the other key I think I would tell people—it just goes back to the comment of a lot of technology offers you a lot of functionality, but you only get to 40, 50% usage. And then how do we unlock those barriers again? And it's never just a tech problem or people, it's a combination of people, process and technology and how do you bring that back around so you can really unlock the full value of the technology platform?
There's so many AI platforms and so again, for us, just what we've done is we've gone back to our COE and tech team. And trying to—you can't let every AI application into your business, so you don't want to stifle creativity either. But you do want to manage it slightly.
So what we've done is actually put AI back into our COE, whether it just be—we don't use ChatGPT because of some of the security things. I mean, people use it personally, but we use Gemini, which is closed to our data source. So there's just different things and just need to consider as the explosion of AI continues, which is great.
Marc: I guess there's a good question though. Like you as a CEO, like how do you handle that with your team wanting to move really fast on certain things and then obviously the security concerns and how does it fit into the broader strategy? How do you create that?
Kevin: We didn't have the framework or I don't know if that's—it just kind of came so fast, right? So then you saw, and then now working with our CIO and through our center of excellence in tech and tech steering is how do you set boundaries but not stifle creativity. So I think there's a lot of products that offer you the security you need as a business that you just need to make sure the people have the training, development access to and understand how to use it.
So that's how we've handled it and it's probably a better question for my CIO, but I defer to him on security, which is really critical, obviously. And but that's how we've handled it, is really now starting to tail that back to our COE and technology and working with our CIO and our security teams and making sure we're fully vetted. We were fully vetted before, I mean, there was no security risk, but just continuing that conversation around security and AI and how do you make sure everyone's comfortable with it.
Marc: And I think that's something that I realized earlier this year when we first started to use some of the agent side of AI, we've really only looked at it from an internal operation side. So how do we handle our support better? How do we handle our field service planning, our project management, our internal documentation and knowledge transfer? Between a sales team and an implementation team, so they don't have to sit there for half an hour and exchange all the information you just talked about for an hour. It can just be right there at someone's fingertips. But then we realized that like all that functionality, just using it internally for operations, if we can then take that and package it up for the customer, there's such a big opportunity and such a big unlock there. I think a lot of businesses that might be doing one thing today will in the future be doing one thing plus AI for someone else. Because that plus AI piece multiplies the value of what your core business is.
Kevin: Yeah. I think you bring up a good point. Just foundational and process, where you talked about—it can be used in a lot of different cases, but the integration between various functions. So sales to maybe your—if you have a BPI group to operations. I do think just that document transfer slash knowledge sharing slash—where is this information? That's huge. So when you start to cross the integration between functions, AI is really wonderful when you just functionally silo it. But when you can broaden that, like with that process, that end to end, it's really powerful.
So I think there's a lot of opportunity in that area you mentioned, we're using it for ourselves. You know, we go from, we sell something. We call our business process integration team. Bunch of documents gets transferred. We have our operations team and then we have our account management team. Quite honestly, that comes back around. So using AI to help bridge the gap, reach those gaps, consolidate the information. Yeah, real powerful.
Marc: There's one big misconception that I still hear sometimes and that's, "Oh, the AI made a mistake." First of all, this technology didn't exist six months ago or 12 months ago in this way, and you're complaining now that, oh, it got this one detail wrong while analyzing thousands of pages from some document. But then on the other side, I also think about it as it's all a matter of context, right? Like oftentimes an LLM might spit out an answer that it just misinterpreted what you asked it, and if you just explain to it what you were really looking for. Kind of like a person that you hire onto your team, right? Like if you hire a really smart person onto your team and the first thing that they give to you, there's a mistake in there. You're not going to fire them on the spot. And you're going to train them up and make sure they understand the world correctly. And I think in similar ways, like all these AI products, wherever they're integrated into a business, they need to learn to understand the company that they're working for in many ways. And your team also needs to learn to work with them.
Kevin: Yeah. I think there's, depending on where you're at and what you're trying to ask or solve, sometimes to me it is just another point of view. You've got points of view from a bunch of people who might have a point of view from AI and sometimes it's our jobs to get through that. And I think on other applications, as to your point, as you train or the answer's going to be 99.9%, right? You know, you just move. So I think how are you using AI and where are you using it? And for what is also important—how much you trust—there are, there's like we go back to our earlier conversation, AI does work on foundational data. So what platform are you asking and what data is being consumed?
So I think it's just a balance of where you're at and how you're using it and you know, I used to have the saying you trust, but verify. So, it's just in AI in some worlds.
Marc: We onboarded the CJ sites to the agent capability and we've tracked over the past few weeks. And functionality is basically, you have a question, just send an email to this email address and it responds back to you within minutes with the findings from all the data. And sort of the phases that we go through in educating or training the team. Phase number one is, what do I ask it and how do I ask it? Do I need to use certain sentence structure or certain keywords or something? No, you just ask it like a person. And if you want to start, just ask it, "What can you do?" It will respond back what the things it can do.
Phase two then is, "Oh, it told me something. I don't think it's right." And that's obviously that trust but verify. And then we always say, well, okay, so did you challenge it like—right, just like a person. If a person sent you this report, challenge it, ask it, "How did you get to this number in this calculation?"
Okay, so that's phase number two, and then number three. Now what we're starting to see some people really interact with it daily is just going back and forth on trying to solve problems. Sure. "Hey, I'm looking into this LPN. Where did it move in the warehouse and who touched it? Here are the five records we have." Sure. "Here are the five people that were doing anything near or at this location over the past couple of weeks." And so they're starting to use it like a partner more so than just getting a single answer and it's really a way to just talk to your data and your system and it's so fascinating to observe that there really is no equivalent to that functionality.
Kevin: It ties back to our earlier conversation, I'll do it one more time, but data, information and intelligence. That theme with AI—kind of what your use case there really is what we're talking about.
Marc: This is great. Well, Kevin, thank you so much for taking time here. We're really excited about the future, really thankful for the partnership that we have with CJ and the many things we've been able to build together.
Kevin: Yeah. Great. No, we appreciate Marc, everything you've done for our organization. I've always enjoyed talking with you, challenging the teams, attending QBRs, so that's been a great partnership. Looking forward to the future. Thank you.
‍
