The Deep Dive Podcast: AI's Logistics Revolution

AI' s Logistics Revolution: Unpacking Efficiency, Costs, and the Future of Freight Communication

0:00

Text from audio:

Welcome to the Deep Dive. We dig through articles, research, and all that stuff to get you the important information. Think of it as your shortcut to being really clued in. Today, we're jumping into logistics. It's often, well, a pretty wild world. And we're looking at how this new wave of AI isn't just tweaking things. It's fundamentally reimagining communication. Our mission is to help you grasp the big forces shaking up this vital industry and, uh, show you a really cool example of how tech is tackling some deep-rooted problems.

Yeah. And it's a really good time for this conversation. The pressure on logistics right now is, well, immense. We're talking a serious need for more efficiency, and cutting costs is just huge across the board. And as we'll get into advanced AI, it's moving past being just nice to have. It's becoming, you know, table stakes if you want to compete.

Okay, so let's unpack this a bit to really get why AI is stepping in now. We kind of need to look back at the last few years in logistics, right? Has been turbulent. Think back to 2020. COVID hits. Bam. Global supply chains just, uh, seized up almost overnight chaos. But then almost right away, people started buying tons of stuff online. So, demand for goods just skyrockets, which meant freight volumes went through the roof.

Yeah.

But we didn't have enough drivers.

Exactly. The classic driver shortage problem, but amplified. So, you get this massive capacity crunch and spot rates price for getting a truck. Like, they just shot up to historic highs. It was just incredibly volatile.

And that initial shock, that disruption, it really kicked off a whole chain reaction of challenges. that are still playing out. You know, we went from those bottlenecks in 2020 to then facing serious inflation and just high costs everywhere in 2021, right? Then things started shifting again 2022. The market softened. You even had over capacity sometimes. And that brings us to 2023 into 2024 where suddenly the shippers have more power and the competition among carriers is just fierce.

Okay.

But what's really interesting, I think, is that through all this up and down, carrier innovation didn't really keep pace with what customer needed. So companies fell back on old habits, making tons of phone calls just to figure out where a shipment was.

Just basic visibility, huh?

Yeah, basic visibility. It really exposed some huge gaps in technology. And yeah, APIs, you know, the tech that lets software talk to each other. They've grown, sure, but they often aren't flexible enough for how dynamic and frankly messy logistics can be in real time. APIs need structure, predefined questions, real logistics problems; they're often fuzzier than that.

That lack of flexibility really hits home when you see the numbers. This is where it gets really interesting and kind of stark. Since 2022, get this. Over 4,000 brokerages have closed down.

4,000. It's a huge number. It really is. That's not just a statistic, right? That's thousands of businesses gone, livelihoods, and the ones still operating. Their profit margins have reached their lowest point in two years. It's a direct result of those operational costs climbing way more. Competition and like you said less overall freight demand.

And let's link that directly to the money. It makes total sense why labor costs are such a massive pain point. In 2024, US truckload brokers are spending over $30 billion a year just on labor.

30 billion.

Yeah. And that's a staggering 40% of their total transportation.

40%. If you break down the rest, actual transportation is about 20%, buildings and infrastructure 15%, technology another 15%, uh compliance stuff, 6%, risk management.

So labor is double the next big cost.

Exactly. And it's pretty clear that having all these disconnected systems and manual processes are the big reasons why they create bottlenecks and huge inefficiencies and they drive up those labor costs. But the key insight here, I think, isn't just that these are small problems. It shows a basic failure of the old human-only ways of doing things to keep up with how complex and fast logistics are now. It's like a fundamental weakness.

Picture this. A team member, maybe several, spending hours every single day just chasing shipment updates. They're on the phone constantly sending email after email, text after text. Where's the truck? Did it get delivered? Can you send the POD, the proof of delivery?

Right. The paperwork.

Yeah. Exactly. So, these teams get totally bogged down in repetitive stuff, constant check-ins, typing the same info into different systems, and all that time spent on grunt work means they can't focus on the important things like building better relationships with carriers or figuring out smarter routes or, you know, actually selling.

So, it actively holds them back from doing the more strategic parts of their job.

Precisely. And because these human communication loops need constant manual input, it really caps how fast a company can grow or react if the market suddenly changes. It's like trying to navigate, I don't know, rush hour traffic, using only a map from 1995 and a pager, it just doesn't scale.

Okay, the challenges are crystal clear. The stakes are obviously huge for the whole industry. So, how exactly does AI come in and untangle this mess of inefficiency and cost?

This is where we're seeing a real like paradigm shift in how work gets done. AI agents specifically are designed to smooth out operations, cut down on those manual mistakes significantly, and give you real-time information. They're becoming essential. Think about it. Today, humans do a huge chunk of the work, especially the boring, repetitive tasks. But in this new era, AI takes on that repetitive load, which crucially frees up the humans to focus on work that requires judgment, creativity, strategy, and higher-value stuff. It's not about replacing people. It's about augmenting them, making them more effective.

And what I find really compelling here is that we're not talking about one giant AI trying to boil the ocean, right? You mentioned specialized agents being more effective. Why is that? Why not just build one big super agent?

Yeah, that's a great question. It really comes down to three things. Speed, quality, and cost, or latency, decision quality, and cost if you want the technical terms. Research actually shows that these big general AI models, they slow down a lot when they have to process too much information at once. A long context window, and their accuracy can drop, especially for tasks where mistakes are costly, which is exactly the situation in freight. One wrong detail, one missed communication that can mess up an entire shipment costing thousands.

Okay, so precision matters immensely here.

Hugely. And there was a piece in Harvard Business Review that pointed this out perfectly. It said firms using specialist AI tend to see a faster return on their investment. Why? Because each agent is fine-tuned for one specific job, one outcome. It's not trying to juggle 50 different things badly. So trying to make one giant AI do everything, it ends up costing more in computing power. It responds slower and it's less precise, especially for those high-volume repetitive freight tasks. where you need speed and accuracy above all else.

Okay, so specialized AI is this smart path. How are companies actually putting this into practice? This leads us to LunaPath, which seems like a solid example of this idea in action. They talk about logistics communication reimagined and is available now. Basically, LunaPath provides AI agents that handle different parts of the logistics world over logistics service providers. Everything from getting a rate quote to handling the final invoice. And the key is to integrate with companies already use it.

Exactly. And LunaPath has come up with a really specific niche. It focuses on that tactical, high-volume, often tedious work. Things like track and trace, constantly checking shipment status. And POD chase, getting that proof of delivery confirmation, pretty much tasks where you need someone or something, making calls, sending emails, texts, basically 247. And critically, updating the TMS, the transportation management system, which is like the broker's brain in real time.

So their promise is simple. Cut down those manual hours, reduce the cost per load using tactical automation that you can basically plug in and start using. It handles that foundational but repetitive heavy lifting, which directly boosts profitability. 

So, how does it work under the hood? You mentioned templates.

Yeah, it starts with pre-built templates. These are designed so you can get up and running quickly for common tasks. Think about settle and pay stuff like matching invoices, checking rates, handling disputes, or pricing tasks like finding trucks. Getting ETAs for pickup and transit, confirming delivery covers a lot of ground.

So, it covers a lot of the standard workflow out of the box, right? But it also lets teams customize. They can create their own prompts for unique situations or workflows they have. And integration is key. It needs to sync easily with the company's existing TMS, loadboards they use, even their phone systems, so the data flows smoothly without ripping everything out. And finally, it's about intelligent execution. The AI agents work consistently 24/7, automating those processes efficiently.

Makes sense. And who is this really aimed at?

It's pretty targeted. Freight brokers in North America, operations leaders at 3PL's, third party logistics companies, and the tech buyers in supply chain. Basically, anyone looking for a fast, measurable return on investment.

Okay, let's talk results. If a broker implements something like this, what's the real impact? What does it mean for their business, their bottom line?

Well, let's look at the concrete claims. The big one is labor cost reduction. They talk about up to 45% savings through this AI automation. By taking routine tasks off people's plates without hiring more staff, you cut operational overhead and boost profitability. Simple as that.

45% is a huge number.

It is. Another key benefit is dynamic scaling. Logistics has peaks and valleys, right? Busy seasons, slow seasons. AI can scale up or down automatically to match demand. And often these solutions are usage-based. You pay for what you use when you need it. That gets rid of the stress of staffing up for busy times, and there's basically zero cost for onboarding new AI employees.

That makes sense. Pay-as-you-go scaling. What about the customer side?

That's the other big piece. They claim up to a 30% improvement in customer satisfaction because the AI delivers consistent, high-quality interactions 24/7. No missed calls, shorter wait times, faster answers, standardized communication. Even though it's automated, the goal is reliable responses that feel humanlike, which improves the whole experience for their customers. We actually have a specific example here. Project 44. They're a big player in supply chain visibility. Their story really highlights the before and after. Before Luna Path, they had challenges. Lots of manual work for RFPs, rate confirmation, slow procurement. It was hard to track carrier performance accurately, leading to SLA penalties sometimes.

Mhm. Service level agreements, right? Invoice problems needing manual fixes, delaying payments, and just generally fragmented communication. Different systems, different teams, carriers, LSPs, All trying to connect - classic communication silos.

Exactly. But after implementing LunaPath's AI, they saw big changes. They automated things like resolving issues, scheduling appointments. This apparently improved the accuracy of the real-time visibility they provide to their customers to reduce the operating costs for their carrier support team. And crucially, they could shift those human resources to work on more strategic stuff.

And Jett McCandless, from Project44 summed it up well. He said, and I'm quoting here, LunaPath improved our data accuracy by enhancing existing processes. That quote really captures it. Better data, better performance, less manual burden. So, wrapping this up, the main takeaway seems pretty clear. Logistics is definitely at a turning point. Tactical AI agents like the LunaPath example are stepping up to handle that routine, high-volume work that used to bog everyone down.

Yeah. The grunt work.

Exactly. And that frees up the human teams to focus on selling, relationship building, strategy, the stuff that really moves the needle. The promise is compelling. Lower costs, way more efficiency, and happier customers. All thanks to targeted AI.

It really is a powerful combination. And maybe a final thought for everyone listening. This isn't just about trucks and warehouses. Think about your own work. Maybe even your personal life. Are there repetitive, time-sucking tasks you deal with? Imagine having a specialized AI sidekick just for that one thing. How much mental space? How much operational bandwidth would that free up? What new possibilities could that unlock for you if you weren't bogged down in the weeds?

It's something to think about.