Intelligent Freight: How AI is Delivering Freight Value | Episode 67

Hi. Welcome to Tire Tracks, presented by Banyan Technology. I'm Matt Silea. Today with me, I have David Bell from CloneOps. David, how you doing today?

Doing great, Matt. How you doing? Happy New Year.

Not too bad. Happy New Year. How's the weather by you? You're down in Florida, right?

I'm in Fort Lauderdale. It's nice, man. It's been a really good weather.

Yeah, so for those of you who don't know a little bit about you, you want to do a quick intro of where you come from and how you got to CloneOps?

Yeah. Cool, man. Basically, in 1993, ’95, I got into logistics and branched off my own business in ’95, and built a really cool low-deride model. That was back in those days, they had a company called KTL was doing service from Florida to California, Friday to Mondays. I love that niche service. I piggybacked off of that and I ended up building that up and became one of the largest partial consolidators in Florida, California and Northeast. I sold that company in 2018. 2014 I started a company called Lean Solutions that became my revolutionizing labor arbitrage near shore labor and transportation logistics. I'm still a shareholder there. Lean has over 10,000 employees in Latin America, Philippines, Guatemala, Mexico. Super proud of what we did there. Essentially, revolutionized.

Nobody was doing it. We were the only ones doing it at 2014. Now a lot of people are doing it. Everybody's doing it. It's well adopted. This AI came out, voice agents came out early last year and I was like, wow, that's very interesting. Another disruption in the industry. I decided to jump in. I was on from Lean in March, or April last year and ventured off to start the CloneOps venture, which is what I've been doing since then.

Awesome. Awesome. Yeah. The AI side of things has been super fascinating and I have been in logistics in the freight industry for about a decade and a half now. It's funny, because some sections of the industry adopt the technology really well and others are not so much. It's interesting seeing – I feel like, the adoption of AI has been met with more open arms than I think previous things. I remember ELDs and tracking devices and GPS devices.

As a former operations manager, we had guys putting tin foil on those things in their trucks to try blocking out signal. I feel like operations has accepted AI a little bit more. I don't know if that necessity or what, but it seems to be accepted a little bit more than some of the other initiatives.

Yeah. I mean, it's still being questioned and still being carefully pursued. I think people are starting to see some benefits. It reminds me a lot of the near-shoring and I think what happened with the near-shoring when we started in 2014, nobody really wanted to do it. I need my people here. It doesn't really make sense. Then all of a sudden, we got some of the big players to adopt it and say it's working, it's a competitive advantage, it's saving me money, it's helping my business and it's good for everybody overall.

It's kind of a lead-follow. Once some of these big players started doing it and getting value out of it, everybody started paying attention. I think that's where we're at with AI. We're right there, where most of the companies that have been looking at it and been waiting and wanting them to dabble, now we're realizing, “Okay, it's time. I have to do something.” We're feeling it.

Yeah.

For the first six months, eight months of last year from from March to October, November really, nobody was doing it. It was like, okay, we're trying this use case of this, and very hesitant. But we felt some time in middle November, December and now, people are ready to go and we're super excited, because I believed in it, just like I believed in near shore. Because when I did it for my business, it worked great. I had a broker team trucking company. I had 50 employees originally in Columbia before I launched Lean Solutions. I think this is the same thing. I'm seeing it working and I'm seeing the people going, okay, this is going to work. It's here to stay. Let's start looking and getting into them. We got some really cool stuff we're doing, just besides the voice AI agents doing carrier or sales stuff.

We jumped into some really cool products, some proprietary products, and some white label opportunities for people that want to own the platform, control it, build themselves and not be stuck in some long-term contract paying per minute for every use case, and then be stuck in a queue with their vendor waiting for them to do it. Everybody's been down the path where they have a TMS. Reminds me of the whole Mercury Gate days. I don't know if it's bad. I better said if I could say it, but when I got Mercury Gate in 2008, I thought it was the greatest thing, until I started having to customize it and wait and the cost.

Yeah. Yeah.

When you take all of the TMSs and you guys, not Banyan, especially, you guys understand, you have a queue and a bandwidth and what you can do in a roadmap. Unlike a lot of the functions in a business in the tech, you have to stick it to your roadmap and your plan. Because you map it out and you can't deviate, otherwise, nothing gets done. It's similar. I think that's where we're at and we're – I mean, it's incredible and the team I have is amazing and we wouldn't be anywhere without the team and the people we built.

We just brought on David Peckman. He's chief commercial officer. Well respected. I'm excited, man. I'm definitely excited about what we're doing with Banyan, too, man. You guys are amazing.

It's super exciting. On all fronts, it's really exciting. I go back to my previous days of being an operations manager in a warehouse and stuff. You touched on this a little bit where people are starting to adopt it. You start to see the adoption in the analytics, the data and some of the things that are, I'm going to say it very plainly, are more annoying to hire for. The first thing I hired for was drivers. I need to get my stuff moved at the end of the day. How I have a customer service person answer the phone, or look up track and traces and stuff like that. That was secondary to just, I need to move the stuff first.

I think the adoption that you're talking about and the timing is actually interesting when we start to talk about it a little bit more is that November period, you start to hit peak season for some folks and they're going, “Yeah, I'd love to have a voice agent just handled this stuff for me that I don't have staff to do. I need people on the road doing things. I need my operations manager coordinating things.” The last thing that comes to everyone's mind is, “Yeah, I need someone to call and do this.” It's like, everyone's busy worker, or the pain be like – That's not the focus. The focus is I need to get it there on time. I need to not damage it. I need to do this, this and this.

It's, it's interesting just how yearly events, like Peak and stuff like that. I come from the parcel space originally. You start to see that adoption not only on the analytics, but the voice agent side of taking over some of that annoying busy work that you normally have to hire for. You can have your team doing more valuable tasks than just picking up the phone and handling, where's my freight?

Well, also, I think too, there's a lot of work that goes into it, or the workflows. If you take track and trace, for example, right, in your TMS, you got, you can put updates and it gives you, and you can email your clients and tell them what to update and you can tie everybody in. But it's always good as the information getting in it. Then you go out and you try to get the macro points and the text locations, and then you go to a nearshore call center to get people to call, and you still don't get a 100% tracking.

What we've done is we built a marketplace where you can basically hire a track and trace agent and have them do the frontend stuff, which is email the dispatcher and ask where the driver is, text the driver and ask them where and tell them to click the link. We have what's called the WYA, Why You At link, click to Where You At link and then bring the GPS back in. Then pick up the phone and call them and say, with the AI and say, “Hey, we're checking in on you.” Because what we have is our AI checks his location and where he's going and what time he needs to get there and tells you if he's going to – if he's not going to make it just based on mapping routes.

When we get that anomaly and it says, okay, he's not going to make it on time. This is where he's at and this is where he's got to be there. Then it calls him and says, “Hey, are we reading this okay. We’re showing you that you're in Southern Illinois, so you need to be in Chicago in two hours. That's not possible. What's going on? Why are you late?” Okay. Yeah, that's right. I'm late. I'm going to need to reschedule it.” Then the AI says, “Okay, hold on. Let me get you over to a team member and work that,” and then gets transferred to a human to jump in, because it's a critical escalation now. Or even better, on our platform, someone can be – that can bring that conversation up to the top of the queue. They can listen into it. Here, with the drivers telling them why they’re emailing the shipper already, have the AI working through the, what's going on and when you're going to be there, why they're already being proactive, or even have the AI email and to do it.

It's a life cycle of a load that you're putting AI in front of to do the things that a lower-level human has been doing either in a nearshore call center and entry level position in the US. That's what we're doing. We're taking all of the work that gets done every day by people and we're mapping it out. What we've done is we have all these activities equal a task and the activity – the cost for activities with AI is a tenth of what the cost is for human. Whatever working minute we can get into the AI reduces your cost by 50%, 75% from a human doing it. That's what we're trying to do is layer AI in with humans and lower – instead of having to add two more people to cover workflow, you have the same two people and AI picks up the slack. It costs you a little bit more for a working minute, but you're paying $4 an hour per working minute versus $18 an hour.

It's an FTE reduction at the end of the day. You're not taking jobs from people, but you're still reducing FTEs at the end of the day, and you're improving the productivity. What's great is like you said, what I see is that's the best use case. I see a lot of people with the best success, where they're mixing AI with the human functions as well. To that same degree, and you touched on this a little bit, how do you see the collaboration of AI with multiple data inputs powering the AI? Do you see that as a benefit? Is that a constraint? In your experience, if I'm using macro point McCloud and these other sources to possibly feed an AI agent, do you see better results, or are less desirable results with that?

No, no. That's what makes it so powerful. As long as you have an API connection to any of these data points, it gets it and it can get it for 100 loads in one second. You can hit all of these data points and you can have a result and a response and a solution, direction into a solution in a second, for 100 loads, or 1,000 loads at a time. I mean, even if you just take, for example, the highway screen, we have a highway screener and we have my carrier portal, verify carrier, whatever onboarding platform you're using, our carrier screener answers a call, pings that, takes the voice, we have a proprietary voice ID for preventing theft, and a lot of people just saw that lobster load that was taken on LinkedIn on a $400,000 lobster load. We actually got the voice form from the company. We ran the voice against our database and that voice matched as a bad actor in our database. We would have caught it.

I feel terrible, because it's after the fact. But we're trying to get it out there and say, “Hey, for less than the cost of a minimum wage employee, you can have this layer of fraud protection that adds to all the other layers and catch this fraudster.” If this guy that stole that lobster load calls a brokerage tomorrow and steals another load and they find out that they could have blocked his voice, I mean, that's where the power is in some of the stuff we're doing. It's all about leveraging the speed in which AI can respond and handle problems to whatever degree, and then having the human have the ability to receive the escalation to know he has to do something, where things –

Where the balls get dropped is when a human is supposed to anticipate, verify something happened and then go fix it. They missed this. They're like, “Oh, wow. I missed that. I missed that email. I didn't check on that truck.” I remember the broker says, I go in and go, “Hey, where's this truck at?” “Oh, let me call him.” I go, “What? He's supposed to be there an hour ago.”

What do you mean you’re going to call him now?

What do you mean, “I’m going to call him. What are you talking about?” I remember, one time, one of my track and trace guys, I came in, I said, “Where's that truck?” He goes, “Let me call him.” I go, “What time is it?” He goes, “8.30.” I go, “What time he's supposed to be there? 8?” I go, “Why didn't you call him last night and make sure he was there? Do you know how important this load is?” He goes, “Oh, I didn't call him.” I'm like, “Listen, pack your stuff and go. How did you go to sleep not knowing that driver was sitting at one of our most important customers delivery? It’s 8.30, and I expect an email from you at 8 this morning telling me he's there. We delivered. Everything's great. I'm going to ask you, you're doing that. Hit the bricks. You got to go.” That's the sense of urgency.

If you have AI that's going to be that person, then instead of waiting until 8.30, I got to come on-offs and ask, at 7.45 a is going to be going, “Listen, I need to check on this. We need to find out what's going on.”

Right. Right. Well, and that's a really great example, because a phrase I always use is if everything important, nothing is important. But in the world of logistics, everything is important. Everyone's shipment is actually important. What's interesting about AI is you can actually action on that, where everything is important and the AI is handling some of that for you. Then it's getting to a human when they need to interact, because yet a great example, you mentioned it. It's like, yeah, at 8.00 last night, you could be dealing with a different escalation. You could be dealing with this, this, and this, and then you go to your email and you're going, “Ah, I should have looked at that.”

It's taking some of that off of there and you can now look at every little facet of your life cycle of a shipment and your operation with a lens you just couldn't do before. Your lobster examples, they’re a really great one. It makes me think about trust in the industry as well. Yeah. I've been around a while. I've seen a lot of different technologies adopted and I've seen a lot of people push back on adoption. Do you think the more of these stories come out about stuff like that, where you're able to use AI to help validate and prevent some of these issues, do you think trust is slowly being gained throughout the industry?

Man, if these fraudsters would have been around – I sold my brokerage in 2018, my trucking. If they had been around in 2008, 2012, they would have stole every single load I had, right? Because the guy calls, “Yeah, I’m your driver. Call me. Hey, it's Joe. I'm picking up the load over there. Baker Norton.” “Okay. Yeah. Go ahead, Joe. Here's your pickup number.” I mean, that was it. Joe got there. Joe got loaded and Joe called me and Joe delivered a 300,000 load of Ramipril blood pressure medicine, and Rancho Cucamongo on three days later and there was no issue. You know what I'm saying? Now, you don't even know if they’re stealing your load. It's crazy. It's crazy how these fraudsters came in so fast quick and so quick.

So sophisticated.

That's the problem with logistics, right? That's so fragmented. It's so spread out. Everybody's doing things so differently. It's hard to get them. Everybody doing things the same. We're working on a bunch of cool stuff. We're working on when a driver gets a pickup, having QR codes go to him, so that they can verify. The shipper just takes his QR code and his load pops up and says, “Okay, this is a load you're getting.” We're working in the community, because it's really near and dear to me being in trucking for over 30 years. I know the pain it is when you wake up and your customer's load's been stolen, and your business. I had to raise four kids on my business and I had so many things happen. You're on this verge of losing everything you have.

Yeah. Yeah.

It's like, to me, there's nothing worse. I want to do something about it and I want to – our voice ID, I think is revolutionary to it. All the financial industries have been using it, right? When you call America, America Express, Chase, they're verifying, “We're using your voice for verification.” It's not that the technology itself is there. The way we're using it in the proprietary way and how we put it together with our database and what we're doing with it is what our patent is about. I mean, people are looking into it and using it, they're crazy, because it's literally, if it saves you one load over three years, it's still three times what you have to pay. Because that load of lobster, I guarantee, that guy's got to write a check for it. There's no insurance cover.

Oh, yeah, yeah. There’s no insurance covering that. You see it so many times, people will wave, or pass on the insurance, outside of doing high-value shipments all the time. Those guys know what they're doing. But you got a couple of these one-off high-value shipments and you're like, yeah, whatever I shipped last week was totally fine. It shouldn’t be an issue this week. That's the one you're cutting a check for and that's going to hurt everyone's wallet at the end of the day.

Yeah. No, I mean, listen, the business is – Obviously, I went into the near-shoring with Lean Solutions. I built that company up incredible. The margins were great. When you look at the margins you make in brokerage and especially trucking, you have no room for upsets and mistakes. There's no room. You literally work on thin, thin margins. If you have one claim, one upset, one issue, there goes your profit for a quarter, you know what I'm saying?

Right. Yeah.

You look at some of the companies, I think there was a 1.3 million-million-dollar Amazon data racks were stolen from a customer. 1.3 million dollars, if you have to write a check for that, and you're going to have to, because there's no insurance that covers that. You don't actually have to write the check for it, because you're a broker. But if you want to keep that customer, you're going to have to, and they're not paying you what they owe you. It's your own claim. I used to see claims come in and my CFO would be like, “What are we doing with this?” I'd be like, “All right, how much business do we do with them? What's our margin? How long is it going to take me to make my money back? Then is there a chance another load gets stolen between now and the time we make our money back?” Because otherwise, what's the point? Let's just settle it and do the best we can and walk away, because otherwise, we might just not do his business and send him a check every month.

Right. Right. Well, yeah. That's a great point, too. It's like, yeah, you lose when you start having those situations come up, then you start running into churn risk. You get these high-value shippers. You get those. You got to hold on to them. You start to have churn risk and that's hurting you in the annual then. You lose a high value shipper, and that hurts your bottom line. Obviously, you pick up a couple of things on a load board here and there. That's different. It's easier to make that up. Yeah, you start talking about relationships and churn across your client base. That's a tough thing to come back from. We're talking a lot about exception handling, how to react and help prevent some of that. On the planning side of things on the very front end, where do you see shippers and 3PLs and carriers utilizing some of the AI tools out there to help planning on the very front end of things?

Yeah. I mean, that's been going on for a while. I mean, if you take a company like Optum, we're a white label partner with Optum. They've been around, I think, since 2008, maybe, or earlier, 2002. They got a great product. They're primarily in the LTL space, but they've now launched their load AI product, which we’re the voice engine behind the load AI. I mean, when you look at them, they've been optimizing freight for a long time. I don't know that the actual AI itself, how we're doing it with the voice and the communications and the AI, like in the planning, but the agentic AI and the way you can have the data now and tap into it combined with AI, the planning is like an no brainer. Because if it's solved, then AI can do it.

We used to do what we call load to ride. I load about 150 trucks a week out of Florida, California each. I'd have 120 trucks going out of Florida. I'd have 100 trucks going out of California and 50 going out of the Northeast on a Friday. We would sit there with spreadsheets. We had every load in the spreadsheet. We had load builders that would pull the load up and pull it, and then build a truck and then get all the freight spread out and then look at stops. Okay, how many stops? Six stops, four stops. Look at the revenue on the truck. We would have to go through and all – we'd have 10 plan – all planning these loads. At the end of the day, everything somehow worked out, or maybe you'd have 10 skids that had to stay behind, or you sent one truck out and you had a loss.

AI, I look at how I used to build that load to ride trucks and my teams used to build the load to ride trucks. AI could do that that now, because we would dump – every truck we ever built, we could dump into AI, going to wherever area and AI can go through now and optimize all of that stuff. AI can do it.

It's crazy.

It's definitely making it way, way easier and more accurate to do.

Yeah. Well, going back to my route planning days, the same thing is, I would spend hours with my station engineer building out zones based off of service level commitments and stuff like that for every single route. It does an adequate job of putting together. But then, your route gets generated and you end up spending half of your morning correcting what the engineer plan was going. Yeah, that driver can't handle that much. Their stop’s for an hour. They're not going to be able to hit that. The STEM times too long, whatever the case is. You spend half your morning just correcting what an automated system is where, yeah, like you said today, I can feed it all that same initial data and history and it's going to optimize it for me like that.

It's going to be able to tell me, “Yeah, you're going to be able to make your service commitments.” Again, I'm from the parcel site originally. You're going to be able to make your service commitments by 10.30 with this money stops with a 95% guaranteed rate. Then you start to see that cascade through the industry, obviously then as well. Yeah, it's super fascinating to see some of those things evolve and be utilized more universally, and going away from the old school spreadsheet mindset.

I mean, if you've read Brad Jacob's book, How to Make A Few Billion Dollars, when he did waste management, that was his whole use case around waste management was coming in and optimizing the routes of the trash. I think he went from – he got 60% reduction in routes by having optimizing the routes of garbage pickup. He was able to go in, instead of having 10 trucks do an area, he had six. He saw that, and even though it was archaic systems they had back then, he was able to come in and upgrade that to be good enough to get that massive 40% reduction in costs in every route. That was his excitement going into doing the waste management roll up.

Yeah. Yeah. With how AI and the learning models are evolving, how the industry is using and adapting APIs more broadly than they were 10 years ago, how are you seeing a normalization of data? Are you seeing using AI through the different data means is pretty normalized today? Or do you think there's still some normalization to go to make it easier for AI to consume these different data sources?

Yeah. Niche and specialty isn't big on the data for what I know. I mean, I might be talking out of school a little bit with my experience when it comes to data. I think data is what everybody looks at as today's gold. Obviously, and oil. Today's data is the oil. How people are using it. As long as I've been in trucking and logistics and maybe people say I'm just dumb and don't know what I'm talking about. But I don't know if the data is all that important, man. People talk about data, data, data, data. If you're a volume broker doing thousands of loads a day, yeah, the data is important. But that's not the market. Most of the market in brokerage, the top 200 brokers represent whenever amount of the freight. You have all of these other brokers that represent the SMB and middle market that are doing 5 million a year and revenue to 200 million. They're just moving freight and maintaining relationships.

I don't know that the data is that important that – I don't know if it's – I mean, getting the data speed and getting it is way more improved. You can have an AI agent be a data analyst now and you can ask it, “Hey, what's my load count last month? What was my market?” You don't have to run the report and do the math. You know what I'm saying? That data points of what you run your business by every day, the KPIs you're getting on a daily basis, I don't know that there's so much more data that makes those KPIs so much more robust than it's going to be life changing for your business.

I think, we all like to say data is so important and data is this and data that. For some, maybe. In healthcare and saving a cancer survivor and going through a dumping millions and millions of scans and being able to find a tumor that a human eye can't find, there's data to process that. In trucking logistics, we're really moving freight. The whole supply chain, maybe data is important when you're going from manufacturer in China all the way to shelf –

Right, right, right.

- in Walmart in Arkansas. I think that's important. But that's not really the meat of the business we dig into every day. We dig into brokers and carriers and shippers that are just trying to get their freight move, track it properly, get the POD, get a bill, get paid. I don't know. To me, okay, data is important, but give me no data today and I'm still going to move all my freight and I'm still going to bill it, you know what I'm saying?

Yeah. Well, yeah. I mean, we've both seen it through our lives of you talked to some of these shippers and they're like, “I've got no TMS. I've got no WMS. I'm doing stuff on spreadsheets,” and they're managing large portfolios of business with no tech stuff.

By the way, if you take those spreadsheets away and try to get them to use AI, nothing will get shipped. You know what I'm saying? It's like, try to pull a spreadsheet away from an old school operations person and they’re lost. You also have to be careful, you don't disrupt your –

That’s the point.

I’ve switched TMSs probably five times over 30 years of logistics. Every single time I move from one TMS to another, it almost ruined my business. It literally, almost –

It's a hard move.

When you're changing operational structure and the way people have been doing things, people are the same. When they go in and they've been doing something, they're creatures of habit, that's a big – change management is the biggest challenge in any technology you're putting into a business. Logistics is a special business. I know there's a lot of Silicon Valley companies and there's a lot of companies out there. When I got into this, few of the top brokers that jumped on my cap table and became immediate use cases were like, “Dave, this is perfect timing. I'm glad you're getting into it, because you understand. These guys from the UK and Britain and Silicon Valley and these pure tech people that have never been in business in logistics, I don't trust that they know my business good enough to understand what I need, because it's not easy.”

Yeah. Right. Right. Really great point. What's fascinating is as we develop different AI tools and we're looking at H1 and H2 of the coming year, not just at Banyan, but other companies as well, you start to go, okay, what's the next AI tool? What's the next product? One that comes up in conversation a lot is doing little things, like tariff-to-tariff comparisons and tariff to quote comparison. You touch on a really interesting thing is you get these old school folks that they want to look at the paper tariff in their hand and there's a lot of value in that. You're going to get a lot of high end shippers and brokers that they work well with that. But then, you're going to have your lower, mid-market shippers, where an AI agent to do it and helps them.

Like you said, just understanding the nuance of the industry and what shippers and brokers and different things struggle and have success in different areas allows you to adapt and apply those products more successfully. Change management is a great point as well, is that being super difficult. Like I said, I talked very early on about having GPS devices in my old trucks and people putting their sandwich aluminum foil on it, because they didn't want to be tracked. Change management, it's a hard thing. As we go more and more down the AI rabbit hole, how do you see more automation and autonomous ecosystems impacting the way shippers and brokers work? Do you see a benefit for them, or do you think the change management is going to be a hard culture switch for them?

I just think it's so big, and so many people doing it so many different ways. Everybody's shipping freight in a different way. Everybody's using TMS in a different way. Everybody does the same thing differently. I think when we look back, it's going to slowly chip away and the way we do things are going to change, and the way the evolution of it is going to change, but it's not going to be from one day to the next. I mean, there's so many shippers, and so many carriers, and so many brokers, and so many forwarders, you know what I'm saying? It's just such a big undertaking to transform an entire industry like logistics and supply chain that it's a 10 or 15-year cycle that we're going to go through, and it's going to be a technology cycle that's finally going to take hold and start being useful and sticky.

Where in the past, you get something new and oh, this is great. They use it and then you don't use it no more. That's because people go back their own way. By the time I do it, I could have just did it my own way easier.

Yeah. EBOL is a great example of that. I remember, everyone's like, yeah, EBOL is the future. So quickly did I feel everyone just abandoned EBOL.

Yeah. Listen, everybody asks those questions and want to know. I mean, listen, you got super smart people out there, like Craig Fuller that are on – now he's a news media guru on his news station and he knows the economy. He knows this way better than I do. I'm just in the weeds trying to build a business, man. I'm trying to build a business with a great culture, with great people, great customers. This will be the third business I've had in 35 years that I've serviced the brokerage community and shipper and community. It's pretty cool.

When I had my low dry company, I haul for every broker in the country. I had 250 trucks, so I was doing back calls for every broker. My service was better than anybody, and they could all count on me. When I launched Lean, they had no problem, because I knew all their shippers. I knew their pricing. I knew their customers. In 30 years, I never had one even instance where someone accused me of stealing their customer, doing something, back selling, or doing anything crazy like that. I had to trust. Now I'm in another business where I'm servicing that same customer base that's trusted me for many, many years. It's really, really cool. I mean, I know you have guys, like Ron Ramsey and Chatterie Lyons, they built a great insurance business off of selling their company in logistics. That's what I want to do.

I think it's pretty cool that I can have the third company, if I could be successful in this company and I'm going to do everything I can to be successful, I think it'll be pretty cool to look back. I got five grandkids and a sixth grand kid on the way. If all of a sudden my grandkids, 18 years from now get a job at a big logistics company and their old grandpa here is a legend in the industry, it would be pretty cool.

Yeah, incredible. I love it. Awesome. Well, hey, as we wrap up, is there any advice you'd give to freight leaders about adapting AI and in moving to this new technology? Yeah, any key advice you'd give them through your experience?

I mean, listen, everybody that I talked to, if you are labeled as a freight leader, you know what you got to do. You're vetting the different companies. You're trying to figure out what to do. It doesn't have to be one winner, one company takes all. I think the biggest thing I'm looking at and what I've learned over the last year is it's going to change. It's going to get cheaper. It's going to evolve. It's just like cellphones used to be. You know when my first business is in 1987 was a cellular one franchise, and cellphones were $3,000 or $4,000 to install in the car, and the phone bills were $5,000 a month. We used to people – guys would install the phone, but not get the service, so they can act like they were on the phone and be cool. I say, you're going to want to get a $5,000 a month phone bill. Now, it's nothing.

AI is commoditized. It's going to be nothing. What we're doing when partnerships like Banyan and other partnerships we have and the white label approach, where we can take an enterprise account like a top 100 broker, and we can give them our platform and they can take our platform, it could be the infrastructure to build all of their AI; their email AI, text, SMS, and voice inbound and outbound and they own it. It's theirs. They own it. They're not locked into some long-term cost per load, no reduction and gain share. It's not expensive, you know what I'm saying? It's not. We're trying to build that and do that.

We abandoned the one-off customer approach about a year ago when I realized, I don't have enough people. I don't have enough money to hire enough people to go out and target these one by one. We did the partnerships with great companies like Banyan who have 5,000 customers that we could tap into and embed our technology into your TMS, so your customers adopt it slowly and use it. That's what's going to happen.

I hope we look back in a year from now and Banyan and we could say, listen, we have 10, or 15 use cases. Our customers will have a 70%, or 60% adoption rate they're using, and sticky and they like it. To me, that's the coup de grâce. That's where we look back and we celebrate, because now we know we're in, we're good. Our customers are benefiting and that's the ultimate goal.

No, I love it. I love it. Yes. Yeah, I love it. Well, hey, David, thank you so much for joining us. Love everything you guys are doing at CloneOps. Really appreciate your time today. Look forward to see what you guys are coming up with next.

Yeah. Thank you, man. I appreciate Banyan and you guys as well, man.