Latest / Rittener Reflections / EP27: Know Your Data: Trade’s New AI Reality

Transcript

EP27: Know Your Data: Trade’s New AI Reality · Rittener Reflections

• Jeff Rittener (00:00) Everyone's talking about AI and everyone's talking about trade. But the real story right now is the data underneath both. Fragmented, inconsistent, and holding teams back at the exact moment they need clarity. If you listen to the last episode with Bas Souti of Sustain360, or even follow the conversations coming out of Washington this week, including the AI plus DC event Axios hosted about AI trust, workforce readiness, and supply chain resilience, there's a theme that's hard to miss. The future of trade is going to be built on data, not dashboards. but actual trustworthy real-time data. And this matters because the pressure on the trade teams right now is unlike anything we've seen. Geopolitics are shifting weekly and sometimes even daily. Enforcement tightening, tariff exposure rising, and executives are asking for clarity at a level that simply wasn't expected five years ago. That's why Know Your Data has become more than a slogan. It's the foundation. Before any organization can take advantage of AI or automation or predictive compliance, they have to solve the basics. Many organizations struggle with fragmented systems, inconsistent classifications, supplier gaps, and the lack of real-time visibility that slows teams down and increases their risk. Today's guest has been living this reality for 25 years. Angela Aaron is the vice president of trade compliance at clearnow.ai. And she's spent her career modernizing processes, building resilient teams, and pushing for clarity in a space that's only getting complex. Angela, welcome to Written in Reflection. Angela Aaron (02:45) Hi, Jeff. Good morning and thank you for having me on. Jeff Rittener (02:48) It's great to have you and you know you've been in this field for more than two decades and When you look at everything happening right now the geopolitics the enforcement the executive scrutiny Why does knowing your data feel more urgent today than at any point in your career? Angela Aaron (03:09) You know, Jeff, that is a great question. And admittedly, I've always been a total trade data nerd at heart. The data itself is the backbone of trade compliance, and it provides the source of truth for what is really going on within your supply chain. That part, it's always been there. But what we are seeing now, amidst the tariff volatility of this last year, along with the global response to US tariff actions, coupled with the sheer amount of data that needs to be mined and managed to wrap your head around it all, that becomes daunting. And especially for leading trade compliance teams, they were already stretched thin. They are now being thrust into the spotlight somewhere they haven't been before to help answer questions coming from their executive teams. They need the tools to be able to deliver on these increasing demands while still keeping up with everything else they were managing before. It's a lot. Jeff Rittener (03:59) well put. You know, for listeners who may not know your full background, Angela, can you share a bit about your path in trade compliance and the work you're leading today? Angela Aaron (04:10) Yeah, absolutely. I started my career in supply chain. I knew that a customs broker conceptionally, you know, imported freight, but that was about what I knew. And back then, I was stuffing NAFTA solicitation letters into envelopes. Not the most glamorous job, but I was curious and I learned to classify HTS codes. I went on to build a qualification process in Oracle. It was a homegrown solution for my employer at the time. and just kept learning and growing. And so I took the Customs Brokers Exam and passed. So I said, okay, well, here's my career path. But back then it was easier to enter the space. Like you could learn at a reasonable pace. And so I just kept learning. was in consulting for 11 years with one of the big four firms. Really got to touch a lot of things, work with a lot of companies. And then what it really came down to when I met the co-founders of ClearNow was a problem. around data, honestly, that the biggest companies in the world were struggling with, and we were trying to find solutions, but everything was just so fragmented. You how do you bring it all together? And that was what ClearNow was looking to solve. So fast forward 25 years, here we are talking now, but data has driven all of those career opportunities along the way. Just how do we use it and empower careers and make informed decisions? Jeff Rittener (05:30) Wow. I love that story and that journey because it's very similar to mine. I started in trade, same thing, processing paper, improving shipments on paper. And I was always struck early on that there's got to be a better way to do this. Got to be a better way to manage the data. And it's just been fascinating to see over the years how technology is continuing to make that easier. And today we're at a point in time where, wow, Technology can really be a difference maker. ⁓ that's fantastic. know, part of your work now is at ClearNow, where you're tackling these data challenges from a different angle. And you're using AI to structure and validate trade data at scale. Now, for our listeners who may not be familiar, how does the platform help teams get their data into a usable state? Angela Aaron (06:24) So we start with something that we call intelligent document processing. A lot of times you'll hear OCR. That's a technology that's been around for many years. That's the process of taking data from a document and extracting it. With IDP, with intelligent document processing, it's taking that data and digitizing it, but it's also contextualizing it. So you know what the data is, and that empowers being able to run reporting. or to automate exception management and use your data in a more meaningful way. It's not just I took a PDF, I made it Excel, but this is my commercial invoice, country of origin, HTS code, unit value. What is the data on that document, putting it into a data warehouse where the sky's the limit on what you can do with it. Jeff Rittener (07:13) That's great. That sounds very attractive. So, you know, ⁓ in your career, you said you consulted ⁓ for, I forget how many years, 10 years plus, and you're working at for ClearNow. So you've worked with lots of organizations across probably many different industries. So when you walk into a new company, what are the most common data challenges that you see right away? Angela Aaron (07:17) Yeah Yeah, you know, really it's how fragmented and disparate the record keeping continues to be. It doesn't matter how big or small the company is, how long they've been established. It's very difficult to maintain that record keeping. Even the companies with the high confidence they can retrieve their records. Chances are they are in a SharePoint somewhere, you know, it's a PDF document. And so if you have a very targeted question, that would be a data-driven response that theoretically could take seconds if you had the right tools, maybe you have to have contractors come in and type PDFs into Excel for days or weeks to get to that answer. And that's very common. So if any of your listeners are like, yeah, that's me, you're not alone in that. We're all trying to get out of that space, but it remains a journey. Jeff Rittener (08:32) Yeah, no, think that's, you know, in the early days, I remember if we had a ⁓ question about, you know, a decision made or a shipment or something, or maybe the government came in and wanted to look at things, you know, it was very easy because we could go to the archives and pull out these huge boxes and we could just kind of flow through and all the data was right there. But today with systems and in a company, there are multiple systems. Angela Aaron (08:52) Great. Jeff Rittener (08:59) Sometimes it's really hard, like you said, to go figure out things are fragmented and it's hard to say. I think that's really part of the evolution of the technology has made it sometimes more difficult to actually get our hands around the data. But you mentioned that the data is fragmented and just kind of curious, why is it hard to fix this? How does this show up when you look at day-to-day operations? How does this fragmentation show up? Angela Aaron (09:27) Well, and part of it is due to the evolution of trade compliance within most companies, right? We've historically been looked at as a cost center, ⁓ made a back office function. You know, maybe there was a customs audit at one point, things didn't go perfect, there was a penalty, there were remediations that were required. And so you built a trade compliance team, you checked the box, you did enough. But this was not necessarily a strategic opportunity within the organization. where a lot of investment was being made. so, know, trade teams knew what they needed. They were very scrappy and, you know, worked lean, but they needed budget, IT resources, and sometimes they could get one, but not the other. And so lining it all up, could take three to four years to set up a global trade management solution or to really move the needle on something that would have a big impact. And so the way technology is moving now, that's democratizing the access to tools. historically, and with the tariffs too, that's adding a lot more attention to trade. But if we look back five, 10 years, it just wasn't within the art of the possible to really make trade a focus. Jeff Rittener (10:40) I have lived that. know exactly what you're talking about. And it's interesting because, as I said at the very beginning, everybody is talking about AI and trying to figure out how do I use AI. Everybody is fixated on trade because it's in the news every day. And you mentioned tariffs. And so ⁓ as we explore this whole concept of data and how to manage it and we're talking about AI. What does AI actually do for trade teams right now? Angela Aaron (11:13) Yeah. And I work for an AI company, but I have to say when used responsibly, it's a once in a generation game changer. It really truly is. ⁓ There's a lot of fear and hesitancy about whether it's safe to use compliance professionals. We're naturally inherently cautious that we have to be like our job is to manage risk. So, you know, is this safe? What is the risk here? But Customs is using AI for enforcement action. You know, lot of pieces, they're, ahead of the game. And I don't say that to fear monger. It's because AI is a great at logical problem solving and trend analysis. So if the government's doing that to look for problem, you know, importers, bad actors, you can be doing that as well. And, and what I mean by that is taking your, business rules, your standard operating procedures, the regulatory requirements that guide your imports. And how can I add more automation as a tool? Jeff Rittener (11:41) Right. Angela Aaron (12:09) So you think about the tariff stacking with all of the tariff volatility, importers, customs brokers, they're having to declare more data than ever before, and it has to be right. And those changes happen so quickly, it was hard to keep up. ⁓ And then you also can automate consistency, accuracy, and at scale, because human eyes get tired. But if machine eyes are looking at your data and saying, hey, maybe you want to look at this, this and this because you asked me to look for patterns or trends, look at this, you know, versus just trying to randomly select and see what you find. Jeff Rittener (12:46) Right. So in a company that you work with, ⁓ how do you actually get your arms around all the data? Because like many companies have multiple systems. And so what's the challenge of trying to get to all that data? Angela Aaron (13:00) Yes. Yeah, normalizing the data is, you know, always been the challenge. If you're starting regionally, like maybe you're looking at the United States, then Customs has that data available in ACE. It's free. Every importer should sign up, especially with refunds on the horizon. That's, you know, that's a big point there. But also if you're using a system... Like maybe you have multiple ERP systems, maybe you have multiple customs brokers and they're sending you your data, but you're trying to figure out how to put it in one place. Then building that data warehouse, you know, with the abilities, the trade ⁓ technology improving to find that one central source and normalizing your data and putting it together in one place. That's something that's been, you know, critical to clear now's infrastructure is that we help companies to be able to centralize that data. ⁓ But, you know, there's many different ways to go about it, but the normalizing the data, taking different formats, different currencies, different structures, and putting them together, that's been a pain point in the past. The technology is really helping to solve. Jeff Rittener (14:14) Wow, that's fantastic. So your tool, ClearNow, how does it use machine learning to structure and validate this data? Angela Aaron (14:24) Well, it's interesting. We've been using AI for trade data since 2018. And 2018 was before ChatGPT was a household name. People were not using AI and supply chain at that point. And our models have changed over the years. When we started out, we were using custom parsers. And now the technology allows for an agentic framework. And we also have what's called a private large language model. which means the data remains confidential. When our customers give us their data, it's not out in open AI. It's not out in, you know, Claude, it's controlled. It's, you know, it's not going anywhere, but clear now. And so as we're using that data, then, you know, the IDP tools, the intelligent document processing, we're creating digital records. We're digitalizing the data. in a meaningful way so that companies can use that data to run automated reporting, to build KPIs and dashboards, develop judgmental audit sampling. The sky's really the limit. And anything that you can dream of, envision doing with your data, if you have it all centralized in one place and the tools to bring it in efficiently, then the answers to those questions can come forward much quicker. Jeff Rittener (15:45) Wow. You know, I like that because I've always struggled with, you know, getting hands around all the data. And then when you do, you know, have it making sure that you can get all of the trade required ⁓ features at the beginning. So now as things move through the process of its journey, whether it's out the door or across the border, you've already got the data there. Right. And so I can see how centralization could be super helpful. So, you know, you talked about ⁓ Angela Aaron (16:02) Yes. Jeff Rittener (16:17) CBP or the government is also using AI and also has access to data and can see where there might be exposure. Can you share some examples of how this AI can help better surface this exposure so that companies can know where there may be issues that they need to address? classification, valuation, origin, those sorts of things. Angela Aaron (16:43) Yes, absolutely. Well, know, and the processes to do this have been along for a long time, but they were clunky. know, 10, 15, 20 years ago, we were sending a Freedom of Information Act request for a CD-ROM, you know, to get your data. That took time to process. And then when you got it back, you better know how to use Microsoft to access. But today, you know, it's all there and available to download in real time for free. And it really comes down to how you apply logic. Jeff Rittener (16:52) Mm-hmm. Angela Aaron (17:16) to mine outcomes from that data. The data is more available than ever, but you can always be overwhelmed by your data. And so, you know, I look at it like this. If you imported the same widget 100 times last year, was the same primary HTS code used? Did you declare the same country of origin? Was your value consistent or were there swings, you know, in your valuation? Maybe you tried a different strategy to reduce your tariff cost. These are things that CBP's AI enforcement is looking for, and you can look for those same patterns as well. And if you're using an IDP solution, whether it's ClearNows or another, you can get to the skew level data from your packing list that may not be available in ACE, and you can compare those. You can almost rebuild your customs entry compared to what was actually declared, or look for those patterns, those changes, those things that might raise an eyebrow in terms of, this could be above board, but let's dig in closer and see if it really is. Jeff Rittener (18:14) Yeah, okay, that's helpful. So ⁓ one of the things that I've often heard and often thought myself is that, trade has become a, ⁓ how would I say this, a role in a company that can really become strategic, right? And really help a company, you know, make good decisions. So thinking about the world today and how we have tariffs, you know, ⁓ being applied routinely, and then we have opportunity to actually get refunds, and you've got a lot of things happening all at once. How does your tool, how can it help ⁓ individuals in companies make more strategic decisions? Angela Aaron (19:01) great question. And it comes down to being able to analyze your data in meaningful ways that model out scenarios. If you're looking to, let's say, resource to a different country, okay, well, what are the tariff impacts of moving to that other country? Or even looking back at the IEPA tariffs have been very big in the news. The Supreme Court determined that the law that was the basis for those duties was unlawful. But we don't have guidance yet in terms of how those refunds will be processed. And so there's a lot of executive teams looking at that right now. How much did I spend in IEPA? What's the material impact? When are we gonna see cuts come back? And those are very difficult questions to answer unless you have all your data in one place. And like with ClearNow, we actually mapped all those chapter 99 codes, all of that, customs brokers speak into plain English. And so you can tell by the program at a very granular level down to the product level what that spend looked like. And so as you're looking at those refund opportunities, section 232 with steel and aluminum, there's something called derivative where you didn't have to pay the duty on the whole value if you could split the metal value from the non-metal value. But that was difficult in the beginning. And a lot of importers paid the entire amount to get their goods into the country. Well, now they're going back and saying, do I have opportunity to seek refunds and recoup some of that spend? And what does that process look like and how material is it to my bottom line? Jeff Rittener (20:34) Yeah, that is fantastic because I believe that the value of a trade organization is to be able to provide solutions to a very complex situation and without data or without having that data centralized or without really understanding what you're actually doing from a supply chain or how you're actually managing the trade data, how could you do that? How can you possibly be successful? Angela Aaron (20:44) Yes. Jeff Rittener (21:04) I think that's fantastic. know, Angela, one of the things that I keep hearing in many conversations, I was at a ⁓ panel discussion at the Hoover Institute at Stanford last week where former secretary Gina Raimondo and former prime minister Rishi Sunak of UK were on stage and they were talking about, you know, how is AI going to ⁓ impact the workforce? And one of the things that was very, very evident in the conversation was that people out in the communities today are really scared of AI. It's like, I don't know if I can trust it. How do I know it's not fake or how do I know it's really accurate? And then there's this sense of, well, AI is going to replace me. And then what am I going to do? These are real things that are going on in people's minds. And so I'd just like to get your thoughts on how do you address those two angles, right? This idea of the trustworthiness of the AI and also the idea of replacement. Angela Aaron (22:08) Yeah. Well, know, AI can mean many things. There's a lot of flavors of AI. It's kind of like going into an ice cream shop. You can go to Chad GPT and ask it a question. That's not an agentic framework. That's not how IDP is using AI. So when I talk about an agentic framework, I mean in terms of automating different interpretations of data, of collecting the data, using the data, but the judgment remains with a human. And there was actually a customs ruling recently, and then this doesn't apply to all industries, but within the trade specifically, there was a customs internal advice that was published that stated, you know, the decision matrix, the assignment of a 10 digit HTS code, that has to be done by a human. The decision process, the judgment has to remain with a human. And so I really view AI as a tool that is going to support people, that it's going to facilitate better quality ⁓ output that richer data ⁓ ease and optimization of making decisions and conducting research. But at the end of the day, the human is responsible. The machine is a tool. The machine is not a person. We're never going to automate imports to where there's not a human in the loop. There will always need to be a human involved in the trade compliance. Jeff Rittener (23:26) That's reassuring for our trade professionals, right? And I think you hit on a great point that it's really shifting the understanding of my role from the person that has to necessarily do all the nuts and bolts of the work, but to a person that actually looks at all that data and makes decisions that are more strategic and more. forward looking for the company. And that's a shift. And I know it's hard sometimes to, when we're used to just doing things, right? And I saw that in my career that there were some people that really just liked to do the same thing over and over. And unfortunately, that's going to be more more difficult because that's what the tool can do for you, right? And you can then take that and make more strategic and thoughtful decisions is kind of how I see that. Angela Aaron (24:15) Yeah, I agree with you. And I think that it brings better quality of life because you're not slogging through data entry per se. You're looking at that data entry to say, yeah, I agree that makes sense. Or could I tweak this and optimize this and really add value versus just showing up and doing a job. Jeff Rittener (24:35) Yeah, and I so wish sometimes that when I first started like you, you know, in the trenches and so on, I was always looking for how can we do this better? There's got to be a way to improve the quality of my life, you know, and so I was always looking if I wish the I wish this technology and these tools were available back then because I would have loved it. It would have been awesome. So but for those of you that have it today, take advantage. This is awesome to be able to have this sort of ⁓ tool and. the access to consolidated and trusted data to be able to make good decisions. think it's fantastic. ⁓ Angela, let's shift a little bit here. ⁓ And I wanna talk about something that's becoming more more important. And that is that as trade and the challenges around trade become more visible and more critical, we're finding that it has the attention of the C-suite. I mean, those in that C-suite are now have to pay attention because there are huge dollars at stake, right? And so when we think about the C-suite, and from your vantage point, what are executives asking for today that they weren't asking for three, four, five years ago? Angela Aaron (25:53) Yeah, you know, it's really a lot of it comes down to the tariff impact because during the previous Trump administration, we had parts of what we see today with Section 301 and Section 232, but it was a discrete set of importers. So if you were impacted, maybe you had some discussions with your leadership, but the majority of importers weren't part of that situation. so, you know, fast forward to Liberation Day. And the average import duty rate into the United States was 1.7%. But Liberation Day said minimum 10 % additional tariff. Well, now we're talking material numbers. A company that had a free duty rate before, you know, they hired a customs broker, they were hands off. But now all of a sudden, if you're importing from China, and you have 145 % tariffs, which did happen between April and May of last year, Suddenly you're questioning whether you can stay in business. I re supply fast enough? I can't build a factory in Cambodia tomorrow. Do I have to shift this product line? Can I still sell it in the United States? And so they're looking for those questions. What is our duty spend? Is any of this recoverable? What strategy can we apply that's lawful, but try to bring down some of these costs? Like what can we do and how quickly can we do it? Jeff Rittener (27:15) Yep. Yeah, I mean, I'm not in the corporate world anymore, but I could imagine, had I been, I would have had many, many questions about when can I get my refund? When can I get that money back? Because I need it, right? I mean, just, I can just hear the questions coming, right? And so I guess what I'm wondering from your perspective, you know, how does a tool lie clear now, you know, where you've got this structured real-time data, right? How does that change the way a trade team communicates the risk and opportunity to the executive team? Angela Aaron (27:51) The access is at their fingertips. And so I hear this every day from customers that there's relief in their voice because someone is tapping them with a question and they can answer it. They're not scared of, how do I tell you how much time it's gonna take to get to an answer that I may or may not be confident when I give it to you? And so that's been huge. But also in the strategy that can be employed here, because when you know by program, by date, by part number, by supplier, you know, if all you have to do is click sort descending and you can see the biggest impact at the top of your list, you know, that that's the game changer. That's if you have a million dollars spent here and ten dollars over there, you're going after the million. But being able to get to the point where you can identify that quickly, that's been the real game changer. Jeff Rittener (28:38) Yep. have you heard any, as you work with companies, have you heard any real stories that you could share about how ⁓ a team has made a huge difference to their executives? Angela Aaron (28:53) Yeah, well, you know, there's an importer that very heavy in the metals and they spent 20 million dollars on section 232 between June and let's say October of last year. And with the clear now tools, they've been able to go through and identify where they spent it and work with their supply chain. And this is a cross-functional, you know, the whole company is involved in one way or another, but they've been able to go back and identify the metal cost and recover those refunds and successfully. Jeff Rittener (29:06) Right. Angela Aaron (29:23) recover refunds and the millions of dollars because they could pinpoint what was going to have the most impact and then, you know, tap finance, tap supply chain and say, hey, I need you to work with this because of this. And it's resulted in real results. Jeff Rittener (29:37) That's awesome. That's awesome. Then you see the real heroes. Right? That's great. Wow. Well, this is, I mean, this is so exciting. As you think about where all this is heading, what excites you the most? Angela Aaron (29:50) You know, honestly, just the advancements in the technology. I thought it was so cool. I was working for an AI company seven years ago and now, you know, six months go by and the technology is outdated. It's advancing so quickly. And so to see that scaling and then the adoption and for trade compliance to have a seat at the table and to have the tools that they really need. And they've always wanted, you know, finally available in a meaningful way that it doesn't take years to implement. It's just a really exciting time to be in trade compliance and it's daunting. It's a lot of work. There's a lot of stress, but it's also exciting. No two days are ever the same and you see the impact of your work every day and you just really can't beat that. Jeff Rittener (30:33) Yeah, yeah, no, I couldn't agree more. I, it's one of the reasons why having left Intel and I'm, on my own. I'm staying in the game because it's like, it's the place to be. It's, it's the time to be here and to really focus on these really, really exciting opportunities. As you said, you know, in my career, one of the things I shared earlier that I always was looking at how to do things better. And one of the things I've always struggled with when I look at trade as it, as it happens around the world is it, it's so paper intensive. And I'm wondering, I've always wanted to ask, I'd love to ask you your thoughts on, you know, do you see a day when we ever will get rid of the paper? Angela Aaron (31:11) I would love that day. ⁓ my gosh, I would love that day. I do see meaningful movement towards being electronic. And at least in the United States, the original copy of a document is required. So if the original that got to the customs broker was electronic, they don't need a warehouse full of paper. They can store electronically on a US server. Other countries, you know, don't have the same requirements. But yeah, we're moving towards electronic. And Jeff Rittener (31:19) Mm-hmm. Angela Aaron (31:41) you know, once we get to electronic, do we move towards something even more seamless? I don't know, but I say the sky's the limit. And as long as we can control the access to the data, the quality of the data, I don't see a reason why we couldn't ultimately get there. Jeff Rittener (31:58) I hope so. That'd be great. ⁓ So let's think about as we get kind of towards the end here, let's think about our listeners and those that are involved in trade. What advice would you give to trade teams who want to start improving their data foundation? Angela Aaron (32:00) Me too. You know what I would say, go to LinkedIn and you know, network really, ⁓ message me, you know, I'd be happy to have a conversation, but there's a lot of tools out there. There's a lot of voices out there. It's a really great time to gain exposure. And you know, I don't work for LinkedIn. I don't mean for that to be an ad, but it's a way to, you know, get access to community. If you have access to any trade organizations, conferences, It's a great way to meet peers and learn what other people are doing. There's more available today than ever before. And in a way that you can turn it on tomorrow, that it's not gonna break the bank, but it can be scary. Where do I start? Like maybe I'm a junior employee. I don't have a big voice in this company, but I'm a digital native and what these tools I'm using are not my world. There's ways to bridge that gap. and it's kind of putting yourself out there and time you don't think you have, but if you're going for a walk in the evening, maybe scroll a little bit and find some things, but the community is out there, we're growing, we're excited to welcome newcomers and ⁓ yeah, don't lose hope, there's options out there. Jeff Rittener (33:15) Mm-hmm. All right. Yeah. Jeff Rittener (33:30) Well, that was great advice. Thank you so much, Angela. I really want to thank you for bringing such clarity to a moment when the trade world is moving so fast, sometimes faster than the systems and the frameworks meant to support it. You know, we opened today by talking about AI trade and the data underneath both. And this conversation brought that full circle. If there's one theme that stands out, it's that data is the foundation. AI can accelerate the work, but it can't replace the judgment, accountability, and experience that licensed professionals bring to the table. Tools can help with efficiency, but responsibility stays with the humans doing the work. These are fascinating times, and conversations like this help all of us navigate them with a little more confidence. Angela, I'd like to thank you again for joining me today. Angela Aaron (34:33) Thanks for having me, Jeff. It's been a pleasure. Jeff Rittener (34:37) This is Jeff Rittner and you've been listening to Rittner Reflections, a forum for exploring the dynamic, complex, and essential nature of cross-border trade, and a space to reflect on the deeper questions that shape how we live, lead, and move through uncertainty. For our next episode, we will dig a little bit deeper into the new realities shaping the trade profession. Thanks again for listening. Talk to you again real soon.