"I joke that AI is like talking to a twelve-year-old with a PhD in vocabulary." — Blake Tablak
During a recent Trax discussion, Blake Tablak, CEO of Trax Technologies, made the case for retiring the phrase "future proofing" altogether. His term is "now proofing." If your organization isn't already building AI capability into how it works, he argues, it isn't behind on some future trend. It's behind now.
The conversation covers where AI creates real value versus where companies are chasing the wrong outcome, the difference between a system of work and a system of record, and how Trax's Prizma.AI platform fits into all of it.
Fresh Off the American Supply Chain Event
This conversation follows Blake's appearance at the American Supply Chain event, where the theme was future-proofing the supply chain. Rather than restate the buzzword, he uses this discussion to define what it actually means in practice: knowing where your data lives, understanding how to use AI as a firm, and resisting the urge to treat headcount reduction as the primary win.
Resources
- Trax Introduces Prizma.AI
- More insights from Blake Tablak
- Trax on LinkedIn
Full Transcript
[00:00] Host: Hi everyone, and welcome to the Trax channel. Today I am here with Blake Tablak, CEO of Trax Technologies. Hi Blake, how are you?
[00:00] Blake Tablak: I'm wonderful, thank you. How are you?
[00:00] Host: Good, good. Summer, always fun. Kids, things, work.
[00:00] Blake Tablak: That's it, the kids are home all summer. Good daycare, it's great.
[00:47] Host: So today we're talking about future-proofing supply chains with AI enablement. Recently you were at the American Supply Chain event, and the theme there was future-proofing supply chain. What does that actually mean in practice beyond that buzzword?
[01:10] Blake Tablak: So I wouldn't even call it future proofing. I call it now proofing. If you're not doing it now, you're really far behind. There are really two things AI has brought to the table, even just in the last year, in terms of acceleration.
If you're an executive running a team, yeah, there's process automation. Process automation has become far more effective. You can do far more complex workflows. And if you're in logistics or transport, you're under heavy pressure to reduce costs. That is the truth. That said, it turns out there's more money to be found elsewhere.
This is where I think the bigger asset is: the rapid ability to analyze information. As an executive, if you're spending all your time thinking about how to cut headcount, you're missing out on the actual big dollars that exist. A good example: one of our clients does half a billion in freight spend. The total number of people supporting that, even if you get rid of half of them, would pale in comparison to looking at service shifts and consolidation and other things, which, by the way, you can analyze in seconds, not days and months, and you don't have to go get a professional and a consultant.
So I think there are two assets. One is process automation, that's great, and everybody should be looking at it. But I'll tell you, the analysis of information and the output is far more powerful, far more important.
[02:34] Host: So what's one capability companies need to build now if they want to stay competitive over the next few years?
[02:34] Blake Tablak: Your knowledge and understanding about how data works would be number one. And two, how you actually use artificial intelligence as a firm. Your data doesn't need to be perfect, but it needs to be organized, and you need to know where it is.
I talked about this at the American Supply Chain Conference. As an executive and a leader, you need to understand the difference between a system of work and a system of record. This is huge when it comes to understanding how to use artificial intelligence.
A system of work is a system where you are doing things, and the information in it is constantly changing. Go to your CRM sometime. Run a pipeline report, run a bookings report, give it a time. Run that same report tomorrow, the next day, you'll get a different result, because things happen in that platform. That's a system of work.
If you go into the ERP and run what you're booking as revenue and expenses, and run that again on a second day, somebody's going to have a problem if it's different. That shouldn't change. Whatever is in your system of record, that's what it is.
Executives need to understand this because they have to know where information is coming from. If you want to know the total landed cost of an object, you're not getting that from one place. There are multiple places, and you need to understand which information to trust.
You don't need to go get a Big Four firm to tell you this. If you've been in this industry for any period of time, you know who has good data and who doesn't. The ability to make a decision as a human being results from your experience, which is data, and the remainder is risk. The unknown presents risk. If you have a line chart and you look at it and say, I have a lot of experience and a lot of data, I can make this decision with very low risk, AI is no different.
I joke that AI is like talking to a twelve-year-old with a PhD in vocabulary. Careful, right? Very sophisticated, but it can be very, very intense about a wrong answer. So knowing where your data is, understanding what a system of work is, understanding what a system of record is, and picking where you want that information to come from and reside, that's the foundation.
[04:58] Host: So where are companies overinvesting versus underinvesting when it comes to preparing for the future?
[04:58] Blake Tablak: Number one, I think companies are overweighting toward headcount reduction. I really think they are. Go back and look at Kaizen. Go back and look at all of these things that still exist today, and what you see is quality and throughput. The two most important assets we have are quality and throughput. If you can increase throughput and increase quality, inherently you can reduce cost.
But if all you're focused on is automating workflows, you will do that, but you'll miss that the person who isn't doing that anymore, who you may have let go, had an asset, or the function had an asset, that actually improved performance, and now that's gone. We see this all the time. "I'm going to use an agent to get rid of this workflow." We actually don't talk about it that way internally, even with our clients. What we say is we have agents constantly monitoring things in the background, making decisions for you, but focused on quality and throughput. Ultimately, a bigger pipeline and a higher quality asset coming through will result in lower cost on your end.
But I think people are way overweighting toward headcount reduction. Once you realize quality and throughput are way more important, you'll find opportunities for operational efficiency and headcount reduction. But first, it's how do I get the object there better, faster, cheaper, intact. That's the number one thing we all get paid on. I think that's where people are overweighted, and it's a little bit dangerous.
[06:18] Host: How should leaders be thinking about balancing efficiency today versus resilience for tomorrow?
[06:18] Blake Tablak: That's a good follow-up. There's a really good opportunity to take on more as an individual. With artificial intelligence, I can actually take on more work. I can do more innovation. I individually can do more as a person. That's really different.
When you think about resiliency and efficiency, quality and throughput again become a great measure. Speed, cost, quality, effectiveness, and how much more of that can I do. If I had ten people today and I was solely focused on cutting five to save a million bucks, great, awesome. But what if those ten people could do the jobs of forty people, and all the other stuff I couldn't get to before, now I can? The result is a hundred-million-dollar cost reduction. The result is a wildly more efficient supply chain that moves goods from A to B faster. I can generate more revenue, and faster, because I can make decisions quicker. That's every CFO's dream: teams that can operate that way. When you start balancing resiliency and efficiency, the quality and throughput measure gives you both.
[07:54] Host: So where does AI play the most practical role in future supply chains right now?
[07:54] Blake Tablak: It depends on who you are, obviously. I'll talk from the logistics side, because that's largely where we spend our time in transport. If you're in logistics, you're under pressure to reduce cost, but you're more under pressure to deliver on time.
What I'd do is pick two or three use cases that seem impossible to get your head wrapped around today. Service shift is a hard one. You start looking at all the packages you're sending all over the world at different points and times. Am I using the right service level to get there? It turns out if you had somebody on your team spend a couple of minutes, you can probably figure this out. In our platform, you literally click a couple of buttons and it tells you: these services between these zones, don't bother using them, because they get there at the same time as a lower-cost service. There are things you can figure out today that achieve the end goal, which is get it there fast so I can recognize revenue, but at the lowest cost, the most effective rate possible.
Another good one is thinking about your carriers, and other carriers. Most of the time, procurement is only limited to what we can see. I know about these carriers, so I'll bring them into the RFP. But there are around fifteen thousand unique carriers in our platform. There's no way you know all of them. The question is, is that data set available, and can I start thinking about it in terms of cost, quality, speed, and efficiency? Who's best at this lane? Who would be best at this leg of delivery? Where do I have single carrier risk? All of that is really easy to answer now.
When you start thinking about AI, start at the end. What's the goal I'd like to accomplish? Don't start with "it's going to be really complex and I can't do it because of this or that." Don't worry about that. Start at the end, and you will figure out how to get there. It needs to be on a coffee mug: start at the end.
[10:06] Host: So how does Prizma.AI support the idea of future proofing? What does it enable that wasn't possible before?
[10:06] Blake Tablak: When we first started thinking about what we wanted to do as a company, and this predates me, it was all lanes, all modes, all currencies, all countries legal to operate in. That was a founding principle of the company. I'm grateful to the founders of this firm that they made that decision, because it's hard if you don't start there.
The reason we call the platform Prizma is that in a prism, white light comes in and a great array of color comes out the back end. We're ingesting plain documents, like bills of lading and invoices. What comes out the other end is a constant set of adjustments you can make. Simple things like: you're using overnight AM and PM service between these locations, but stopping at PM gets there at the exact same time. Nobody else is doing that for you. The platform is doing that for you. You've got single carrier risk in these lanes, here are twenty other carriers that do that at better quality and potentially lower cost, you should consider them.
Prizma is designed to give you a different view of the world. I tell our executives: don't look at it and say, well, I don't see it, it's not in there. One of the great parts about Prizma is that whatever question you have, the data can be constructed to give you the answer, and then you monitor it.
It's also designed to export its information into systems of record. We are a system of work, and I'm very clear about that with our executives. Your inventory systems are systems of work. We are a system of work. When you combine that information in a system of record, that's where you start keeping track of historicals, and that's where the real power comes from. Now you have an integrated system, and that becomes wildly powerful.
So Prizma really does two things. One, it gives you the ability to analyze things very quickly. Two, it moves that data from one system to another in a structured fashion. If you're a large-scale enterprise, that's what you're trying to do, twenty-four hours a day, seven days a week.
[12:28] Host: So the last question: any other trends, insights, or observations you've picked up recently, especially from events or conversations with industry leaders?
[12:28] Blake Tablak: Maybe not specific to recent events, but I'm old enough that I was around for the shift from mainframe to distributed, distributed to SaaS, and SaaS to AI. One thing is constant whenever there's a ton of disruption: if you're terrified and hold on to the thing that you thought made you valuable before, you will get clobbered.
I remember my father telling me, be careful what you ascribe value to, because the thing you ascribe value to, you will protect. Now, there are obvious things, your family, your faith, you should always value those. But when it comes to your work and your output, be careful. If you thought your value was in really understanding how to make logistics work and move objects from A to B, that's changing. The output still has to remain the same, but as a leader, I'd tell you: become more comfortable using solutions that involve data. Become more comfortable in your data, and know what's possible.
Don't let that be a gray area, because your team may tell you something isn't possible, and one of those reasons could be that they don't want to lose their job. They're not trying to hoodwink you, there's just a lot of emotion in it. The more you know, the more you can empower your team, and the more you can actually go do yourself.
My grandfather was sixty-seven when he learned how to code. He'd been an accountant, he'd been a CFO, and at sixty-seven, his son started a software company and he said, well, I'll learn how to code. And he did. There's nothing in your life you can't go learn. This is a pivotal moment where billions, if not trillions, in value are about to be unlocked. If you're on the back end of that, it's going to be awful. But if you embrace it and realize the value you used to provide is changing, you'll become the center of the universe for that company.
I'd really encourage people to stretch themselves a little and start thinking about new, creative ways they might add value back to their organization.
[15:03] Host: Yeah, absolutely, sound advice. I definitely think continuing education, especially in AI, in any field, is paramount to your success. Well, this was such a great conversation, Blake. Thank you for joining me today. We'll make these a regular thing.
[15:03] Blake Tablak: Thank you so much. Great to see you.
Watch the Full Conversation
For more from Blake Tablak on AI, data strategy, and what it takes to run a resilient supply chain, follow Blake on Trax's blog or subscribe to the Trax YouTube channel for future conversations.
