#100: Climate Intelligence for Global Supply Chains - with Jonathan Horn of Treefera

Speaker A: Hello and welcome to the Leaders in Tech and E Commerce podcast. I am your host Andre Palamario and I am the head of innovation for ELCAD Global and the founder of Supplify. Our mission is to connect the supply chain, tech and e commerce ecosystem globally by bringing forward some of the most interesting stories about success, failure, lessons learned and more from leaders in the industry. I'm um, very happy because today we are going to talk about risk management, intelligence, AI, uh, visibility, so all very interesting topics in the world of supply chain and procurement. I'm happy to have with us today Jonathan Horn, who is the founder and CEO of treeferra. Jonathan, welcome.

Speaker B: Thank you very much. I'm delighted to be here and great to have this opportunity to speak to your listeners. Thank you very much.

Speaker A: Excellent. So maybe just to set the scene, Jonathan, we can start with a quick introduction of yourself and also maybe you can um, tell us how did you start with Treifer, what the inception point?

Speaker B: I'll start with a little bit of background on myself. Let's be mercifully brief. It'll make sense then. So I think my journey as uh, farming, finance and physics farming came first. I was born, brought up on a farm that's become relevant recently and very useful just to know a little bit about how seriously things, you know, how people grow stuff and what the challenges are to farmers and supply chain and real sourcing in the first mile. My second phase was really physics. I was physical research for a long time. So I really got interested in things like fluid uh, flow and artificial intelligence and others. And then I did 20 years in finance. So uh, that's 15 years at Citigroup and five years at JP Morgan. And that was mostly in the kind of space of risk and control and particularly on the risk side at JP Morgan had spent a huge amount of time with other colleagues really understanding assets, the balance sheet and using all sorts of artificial intelligence methods to do that on a huge scale. So it was a wonderful journey and only one that made sense in hindsight. Uh, but how that led to Tree Fur really was. I started thinking a lot about, you know, sort of in 2000, temperatures getting hotter. Uh, from a physics point of view that's, that's not a political statement, it's just a statement of physics as ah, things get hotter. They don't just get hotter and things change gradually. They kind of go crazy non linear really fast. And that is fluid flow in its essence. So what does that mean for all of us? It means that there's going to be More extreme, more frequent and more impactful events from a weather and climate point of view. And that has a huge and profound effect on all of where our food really comes from. So we can see that these effects will be more frequent, more out outside and it's this outside human experience. The only time we've kind of, in my experience anyway recently understood non linearity of events was during COVID So every morning when you got up in Covid land things were like more peculiar than you could possibly imagine the day before. So that's the kind of non linear experience. Uh, the same will happen with the effects of temperature and then the other kind of side of why tree furrow? Why now does it make sense now? So temperature rises, things in the net crazier. That's going to have a profound and volatile, you know, disruptive effect on food. And so the data about where things comes from gets more and more important. The second element then is technology. Satellite technology has improved enormously in the last five years. AI has gone through the roof in the last five years. Uh, we all experience that through large language models, anthropic OpenAI and so on cohere. But the same kind of revolutions taking place in our side of AI which is m much more about signal processing. We take a vast amount of satellite data, radar data, open uh, source data, all sorts of information from places and we use the wonders of artificial intelligence to munch and crunch on that uh, to get something that makes sense to clients in their business domain. That's the whole purpose of treeferra and that's where it came from.

Speaker A: Very interesting to hear about the farming background physics and then finance and risk and now how Trifera comes together so you can only connect the dots looking backwards like you said. So it's interesting. What can you tell us about Trifera when it comes to the idea of what makes it unique? Maybe we've seen so many players uh, being bigger, smaller startups in the visibility, traceability, sustainability, you know, all end to end supply intelligence. But I think what you're building has a uh, quite unique angle. I'm very interested to learn more.

Speaker B: Yeah, sure. And because, and you're right, I mean those people have made some very valiant attempts and some good attempts over the years to try and get better views on what's really happening in the full supply chain. We've focused very much on the first mile of supply chain. That's what's critically important. And we feel that that area is underserved with Midland. Supply chain has been known for years, companies are super crazy expert on this, blah, blah, blah. There's a lot of technology supporting it. Last mile has really been cracked by a lot of distribution networks. So Amazon, UPS and all these other wonderful companies have really cracked the last mile. The first mile is underserved. So in eggs and soft, so you know, grains, so soy, wheat, uh, and then on the soft side you're talking coffee, cocoa, sugar, cotton. In those places and spaces, there's a variety of things. People have used satellite information and do use satellite information and they use, we also use a big network of experts in the field. So people who agronomists and others who are stationed in various different places around the world and provide insight. And then they also, they've got one foot in the river as I call it. So they're typically got a lot of connection with the physical commodity. So in building their view, their worldview of supply and demand, they really understand that that space. What's fundamentally different about tree fur is that we'll spend, we use the advances in satellite and the advances in artificial intelligence to really unearth very granular, near real time way what's actually happening with crops and commodities on the ground. And so regardless of how good your network is and so on and so forth, of course via satellite and with the power of artificial intelligence, you can interrogate every plot in every location around the world for all of the key commodities that everybody's interested in. You can get that in massively granular detail. And the information that you can get from the models, the artificial intelligence models every day, every week that's improving. And even so for tree furrow, you know, we really started operating uh, in 2023 and the progression of models even in the last six months with deep learning models, embeddings, all sorts of other stuff from Earth observations that's, it's made those models night and day different in terms of their capability. So for instance, identifying in real world things cocoa in understory trees. So Coco Coffee and others, as many people on this podcast will know, often grow some out in the open, some, you know, kind of within an agroforestry kind of environment, using signals from radar and opt you a signal, uh, that you can process, you can understand and accurately identify where this stuff's growing and the condition of it with uh, much greater accuracy and precision than you could even six months ago. So it's like, I'll just say to people if you think right, you know, my experience since I use Claude a lot, my experience of Claude in the last, last six months has been revolutionized in a way that you know, would have been hardly credible if you thought about it three or four years ago. But uh, the same's happening in this space. So just think, you know, whatever your experience of running Claude's bin or OpenAI versus in the last six months versus it's, the acceleration is exactly the same. So these incredible insights are coming thick and fast and the ability of these models is improving all the time. I think today it's a great supplement to the data sources that we have. I think over a period of time we'll come to greater, greater reliance on the satellite and A.I. ah, based uh, modeling with less uh, physical on the ground kind of presence that'll become less. I just think it'll be kind of outperformed over a period of time. Not today, but over a period of time.

Speaker A: And when we're thinking or you're looking outside and you have multiple conversations with different type of clients, different geographies, you're dealing with global MNCs. I'm sure there are patterns that come up in the discussions, so to speak. Trends that you can see in different industries or even across industries that are happening connected to what you're trying to solve for. So I was just curious, what are some of those trends that you see across industries?

Speaker B: And so who do we, you know, so if you think about true for what do we do? We do market signals. So that's information uh, about commodities in various different places that people use as a source for trading. So that could be trading futures or it could be trading procuring essentially. And we then also uh, work with a lot of kind of uh, strategic supply chain companies. So what are they trying to do? And it all comes back to this central thing where uh, we're still underserved with information on the first mile when the world was benign, you know, and there wasn't these kind of temperature rises, this didn't really matter. I mean obviously if we go back a long way in time, people are very used to the concept of seven fat years followed by seven lean years and all the rest of it. But it was, you know, the world was much more benign. And so crop failure and food, you know, insecurity, it wasn't as profound as it's starting to become now. So, so I'm seeing a trend across all of these things. There's more volatility in crop commodity pricing and particularly on futures contracts and all the rest of it. Why is that happen? Two reasons in my opinion. Number one, fundamental data is changing about the source. So the supply side in the S and D model driven by the kind of fundamentals of physics and all the rest of it. And then the second thing is because of that volatility and the size of these markets, it's starting to attract a lot more uh, traders that are purely looking for alpha. They're not in the physical commodity business, they're looking for alpha. And that drives its own kind of reaction to these events. So, and that just then gets into a cycle of people need more data, uh, to understand what's really going on in detail. And so the frequency and the cycle and the kind of um, you know, the cadence of this is massively speeding up. And I see that across different industries. Uh, and then kind of on a Mead. So that's the kind of high cycle that's you know, now till the end of the season where that kind of high cycle thinking and then kind of more medium term, uh, what I see is companies, and I'll give you a specific example of client of ours, companies are more and more having to think about things that were kind of not really thought about previously. So they're thinking about where we're forced now to move the origin of our production or you know, the kind of origin of our supply chain from one location to another. And it's not just a episodic thing. It's oh, it was just a bad year, it's persistent, frequent catastrophic years are leading people to have to think, right, we're going to have to actually seriously relocate at the source of where we're getting this stuff from so they can mitigate where it is. There's, you know, there's uh, positives and benefits of doing that. But a good example of one of our clients, they're specifically looking to rethink uh, where they're sourcing their commodities from or a couple of key commodities. They used to get this commodity from, uh, Eastern Europe. They're specifically moving it to sub Saharan Africa. And what they need to be able to do that reliably is they need to think about, of course, the cost model. Super important. What's it going to cost to produce this commodity in that location. But also much more than perhaps previously. What's the risk profile of these new locations? So am I, in fact, am I just stepping out of the frying pan into the fire, or can I scientifically and over the next two, three, five seasons robustly understand the weather and phenology risk to this crop? So what does that mean? It means, you know, on my commodity, am I Getting the next three years, do I have a standard decent chance of not having significant failure to deliver on the commodity? That wasn't, you know, people always had, oh, there's a bit of a shortage this year or whatever. It always used to be that, but now there's, you know, significant fat tail risk of total loss or total failure to deliver. That wasn't, I don't think, uh, that hasn't been a thing for a long time.

Speaker A: I will give you two examples that came across in an, in an event we've done recently in Zurich. And there was a discussion about two things. One was the crisis or potential crisis. Now it's, it's a fertilizer or lack of thereof because of what's happening or what has happened so far in the hormoz trade. And then the other one was, how does this year El Nino phenomenon impact weather plus the ability of ships crossing the Panama Canal because it, uh, rainfall

Speaker B: and all the rest of it.

Speaker A: Yeah, exactly. And I just wanted to put forward these two ideas and just see what's your take and how would treifera be able to help with.

Speaker B: Yeah, so, um, I've just come back, uh, from a business trip to Singapore. We've got an uh, office out there in Singapore. I've had maybe 47 meetings that came up about 37 out of the 47 times exactly those things that you're talking about. So what does that, you know, what is that in term in tree fura terms? Um, tree fura is very good at, ah, modeling the forward this season and over a number of seasons, effects of El Nino. Of course, those weather models have their own intrinsic, um, kind of uncertainties that everybody, uh, knows and understands. But what the, what the platform can do is say, okay, for a given set of commodities, you know, I'm perhaps interested in grains or I'm interested in coffee or cocoa, whatever it is. What are, what's the range of outcomes that I can expect? Uh, so the range of outcomes and if you add on to that, you know, the modeling of say, fertilizer and what the impact of that might be. So people, maybe not this season, maybe people in farms have mostly bought into this season, uh, but maybe next season, that's going to have a tremendous impact. So modeling that El Nino risk in a way that's really specific to a particular crop in a particular location, what we're generally finding is that, um, a lot of these models, and again, talking a lot to people in Asia about this, a lot of these models say, oh, maybe there's going to be like a 5% reduction in yield or maybe a 4%. But actually a lot of the models, when you look, look at them, there's a fat tail risk of things like 30% reduction, 50% reduction. There's some really extreme possible and completely realistic scenarios. So now what does that mean? And it won't happen everywhere. It'll happen in really localized places. So what that really means is you can get to a place or get to a point where you've got a really detailed kind of map of where you're getting your commodities from. You can get a really detailed view of the possible, you know, the ensemble of weathers that are, uh, completely realistic from a physics point of view and that can give you a much greater sense of your fat tail risk. And in general, uh, my view is that and Tree Ferrari's view is that we are overall underpricing risk quite significantly. So if you can get those views at the moment, everybody's, you know, has a gut, experienced people know, uh, they've got a gut feel that this is going to be a problem. But precisely what's the, you know, what's the possible extent of that problem? Where will that problem occur? You know, is it going to be a really extreme effect in a localized area, or is it going to be a profoundly and um, broad effect, broader area? Those are the things that need to. People need to model. Now, uh, before Tree Ferrer, obviously people had these S and D models. What did they do? Very sensible things. They went in and they said, okay, what if it's 5% worse, 10% worse, blah, blah, blah, what will that mean? And you come up with an answer. But if you knew that there's a very high likelihood that it's going to be 22% down, then obviously that changes the scenarios that you focus on in that S and D modeling space. So I think people will start to zero in on the bleaker side of possible outcome funds. Not to catastrophize, I'm, um, not in the business of doing that. I'm just in the business of saying realistically and sensibly, is there a 30% chance that we're going to get a, whatever, 50, 60% hit on yield? If there is, what would we actually do? And that allows you then to take some very straightforward business decisions earlier on. And all of that will fundamentally protect your business and protect your supply chain. So I think it's about, instead of just our normal, let's run through a bunch of kind of typical scenarios. Let's run through in detail scenarios that are uh, likely and are part of a tail kind of risk profile.

Speaker A: And now if we have to think about the client story, Jonathan and I m know there's quite a few of them that did well and if we can give the audience examples of doing well in terms of results that they achieved by using Trifera's intelligence platform, that would be great. So curious to hear some client success stories.

Speaker B: Yeah, perfect. I'm actually going to give two one will be probably less relevant but I think very interesting. Less directly relevant but interesting to uh, your client audience. And then another that will be really uh, super relevant. So I can name this plant because it's part of their website. So that's straightforward. So we provide data to JP Morgan as an example. That data is specifically on beef and risk to beef production in the US and precisely we're looking only at the kind of the start of that supply chain. So we're looking at head of cattle. We're looking at weather and climate scenarios, not just uh, you know, kind of average selected scenarios, but real scenarios that can, are being forecast now. And then we look at kind of the um, what effect does that have? If you look at big beef producing areas in the US Like Texas isn't as a great example, they carry a lot of head cattle. There's a pressure on water that's a prolonged and has been a cyclical but prolonged, gradual, you know, uh, unfavorable effect and that reduces the amount of forage available for the animals. So herd sizes have to go down. There's not enough to eat. Of course that's bad news for the herdsmen, the ranchers, the farmers. That in itself then drive is, you know, drives a price pressure in the U.S. of course there's other factors, there's tariffs, there' all sorts of other factors but it drives a real, you know, it's a really important part of that drive, that uh, information. Then it takes, it goes a long way to explain why beef uh, cattle are you know, kind of finished animals are ah, 20% more expensive than they were kind uh, of last year and the year before. And for certain sectors that goes up to 33%. So that's a real impact that's flowing straight through to consumers. It has its own kind of social geopolitical effects and all of these things are really important and ultimately feed into food security. So that's an example of risk essentially uh, our risk, uh, intelligence around commodity, in this case beef, uh, and kind of our forecasting around that, using that, the detail. So that's a Kind of finance example, maybe less relevant, perhaps some of your audience, but ultimately that, uh, key point here is that banks are starting to use this data to assess supply chains. And therefore ultimately it is very relevant at the particular board level, uh, in these companies just to know that banks are starting to do this. The second example I'd give then is um, around kind of location selection. So if you're faced with, so we've got a whole bunch of crops and commodities, I want to be assured of a certain supply, very, uh, specific client that we've got today. They're looking at, you know, they've got 150 different possible locations that they can select for sourcing of a particular commodity by looking at these risk maps. So again, it's more, kind of, a bit more of a strategic sourcing play and procurement play. By looking at these risk maps. They can say, okay, if I could go through all of these potential places where I'm going to source, and I look at these risk maps and they are essentially the phenology of the individual crop plus an ensemble of possible outcomes. It gives you a spread of yield and therefore a spread of delivery from these locations. Big players particularly have got options about where they go. They can cherry pick that list. So now not only are you making, you know, the right decision in terms of cost, in terms of quality, but, uh, you can also make a decision, a very informed decision based on risk as well. So risk of failure to deliver because you can see from these, uh, outputs out of these locations, if I just kind of took the top quarter, show me the ones with the highest yield, least yield, risk, and then I'll also look at cost obviously and quality as well. So it gives you a very precise answer. And we've got a number of clients, household names in that space, who use this technology for, in that exact example, in the case of corn and in the case of coffee, uh, and cocoa.

Speaker A: Now, coming back to clients, um, Jonathan, that's, that's an interesting perspective that you said, both for the banks and for the risk mapping, so to speak. But I imagine clients always have access to data. Now how is that data managed, how accurate is the data, how is it presented and so on, that's different things. But if you were to think about the most frequent aha, uh, or aha moments that you heard from the client's mouth, especially when they see the data more clearly, I'm just wondering what are the first things that they kind of say or react or the things that change in their mind once they have a better Understanding or of what's the reality. Like not their old model or the old way of thinking about this.

Speaker B: I mean we're lucky enough to have some amazing clients. We deal with a lot of uh, real blue chip household names. And so they're very forward thinking. They've got incredibly capable and smart teams. They as you say, collecting data for years. So they really, really know this space. Um, so they're of course, and that's how I enjoy working with those people because they're always going to check and challenge what we're doing and that makes us better, better. So I love working with those people. The big aha moments and I'll give you a couple of examples. So doing a whole load of analysis uh, for one of our clients and this is all done on the platform automatically. But the big aha moment was they've got a whole load of data on field, say in uh, the Midwest, on soy, on corn and other things. We went away with some of that data uh, which they provided to us to help train and validate what we were going to do. So it was like a kind of view into this space and how would it work? We came back on them. So this was on whatever on the Friday, uh, we triggered a whole load of data pipelines, all sorts of AI and whatever models that ran over the weekend. On Monday morning we had a clear view going back seven years of 4.7 million fields in the US the entire history of every crop that's been grown on that, on those fields, uh, what which are all automatically identified through the phenology and how that shows up in the satellite imagery. So it's demarcation, uh, of all the fields really they're most interested in corporations. Corn in soy in cover crops in till, no till. And it showed a whole kind of for each of these fields, a whole history for the field that they could themselves validate. But what was the aha moment was wow, you can do this at like colossal scale and in really short order. So while you know, expertise humans in the loop will always be incredibly important. The ability for these models to amplify capabilities that the company has kind of in maybe small scale is absolutely amazing. And then it also allows you a, to take big scale decisions decisions. It allows you to run uh, scenarios based on unseen but entirely possible, even likely weather outcomes. So you know, obviously the weather we experience any one year is just a single strand of weather essentially for every location. But if you could run 51 different options around what that weather could have been and um, very likely to have been, then you can start to see that the world that we're living in has got greater amount of uncertainty and you can manage and respond accordingly. That's the brilliant thing about it. You can respond much faster, you can think more comprehensively and you're aware, you know, you're not, you're less likely to be completely caught out by certain kind of types of events. And when the world's changing in a non linear way, that's incredibly important to be able to do that.

Speaker A: To this point. I was talking some months ago with one of the larger chocolate producers in the world and they were so, and I was talking with the person who was looking after procurement analytics and he was saying, look, uh, we are trying to solve for a drought problem that we have across Asia and that impacts the cocoa production and that impacts the prices that we buy the cocoa at. And we haven't been able to predict this. So now we are in a bit of a tough pickle. And um, it's not an easy. And I'm tasked to figuring out what's the best strategy and it's um, it's, it's difficult. So I suppose that if somebody like them would have been a Trifera client, this scenario wouldn't have happened.

Speaker B: No, they'd be able to get data on it. And, and what these capabilities are very good at is understanding the possible nonlinear effects. So if you just go, you know, my experience in life, you know, when I was uh, brought up on the farm, whatever, what happened 20 years ago, 30 years ago is important. Of course it's important, but it's becoming less relevant because we're having, you know, people say, oh, this is the worst. Every year we seem to have a hundred year event, we're going to have to start reclassifying those events. Is, is happening a lot. So it's that kind of thing. And you say, look, there's these weather fluctuations, there's El Nino and there's a long term shift in the precipitation pattern that's happening in coffee in Brazil as an example. Our first response to that of course is okay, well we need to irrigate. But at some point or other those remediations kind of run out and we have to, we have to know, you know, actually if you just look five, 10 years ahead, this is no longer going to be vast, viable to grow coffee. If you look at Hungary and corn growing in Hungary, a lot of Hungary, you know, used to 20 years ago, corn, you know, there's a huge amount of corn produced out of Hungary. It's far less viable than it used to be. And that isn't like a farming problem, it's a, it's a weather problem. Um, so, so yes, this looking kind of more broadly being able to incorporate these very short term effects, the medium firms effects around the El Nino and then these longer term kind of drive dying or increased precipitation kind of depending on where you are in the world, layering that all up together. It's very hard for human beings to do that. That's one of the powers of AI. It's not going to tell us what to do. Well, not yet anyway. But the thankfully, but it's what it's very good at is bringing together very complex sets of signals and then distilling them down into things that are very likely, quite likely or completely unlikely can just be eliminated from our thinking.

Speaker A: Now changing a little bit the registry, um, Jonathan, coming back to Trifera as the company, as the team, I'm wondering from your perspective as the CEO, uh what's your ambition or what's Treifera's ambition as a company moving forward it can be one year, two years, maybe more. And that's uh, one part of the question and the second part of the question is what are some uh, exciting things planned for the product roadmap that we might be able to see in the future?

Speaker B: Yeah, perfect. So I think of it as a joy and a privilege uh, to work at treeferrous. I find it uh, fantastic. I've been lucky enough to work with some brilliant companies and brilliant teams over the years. Tree fer. What is it? It's three years old from inception for a series B capitalized company. We're growing uh, rapidly and I think this whole space is growing rapidly. So obviously I credit myself as the strategy officer of that. But um, that's you know, that's just how things have turned out. So this space is very important. It's very exciting. It brings together for me all the right things. Serious science, serious artificial intelligence, serious engineering, but with a proper commercial outlook. The goal, the real dream of Treefarer, our vision is to influence the flow of capital through data. And by providing people with the right data, the decision makers. I think people can make great decisions for food security, for securing supply chains, for thinking carefully about what remediation might help in the medium and long term around sourcing and farming practices. All of these things are going to be good things in the world. They all need addressing. Uh, having that data to make precise and data driven decisions about them is incredibly important. Our vision is to influence the flow of capital as a result and hence our focus on, you know, financial services and big supply chain companies. That's, that's our natural space. What am I excited about in Tree Fura? Uh, what I would say is, um, it's not for the faint hearted. We've got a brilliant team. They're kind of younger, uh, they're in their 20s, 30s, uh, most mostly. So, uh, they're very capable and believably well qualified and ambitious about what they're doing. And I'm lucky that most people at Tree Fura have actually joined Tree Fair. Not only because it's a super interesting problem, but it's also purpose driven. So a lot of people, I kind of really like that. I certainly like it, uh, I love that part of it. The exciting things that come out every day in Tree Fair is a training day. Any kind of, uh, idea that, oh no, I know this space that's immediately eliminated by day two because everywhere you turn somebody's got a PhD on this and that and the other. Uh, am I excited about, I'm excited about. And it's kind of mundane in the way it sounds, but I'm excited about the Tree Furor agent which we're surfacing now internally and to use now. What's different about the tree fur agent versus, you know, the multitude of agents all in the world? Tree Furor fundamentally makes observations about the real world. That's what it does. Those observations are scientifically robust. They come with a qualification around the quality. So you've got an assurance about what reliance can be placed on them. So it's not just some nonsense that's kind of been churned up from, from some disreputable article on the Internet. It comes from a hardcore decision. So ultimately it's come. The information that Trutharer is surfacing through. This agent is about data that's been observed in the real world with robust methods and with the right confidence. So people can think, okay, I can actually, you know, I now know, uh, these are going to be the outcomes. This is the likely risk. This is what's happening in cotton in western Texas. You can ask it like plain, straightforward, you know, plain English questions, straightforward questions and get back an answer, uh, that is supported by some rigorous data outputs about observations that have really taken place. If you ask that question to Claude or OpenAI or any of these others, Claude is going off to all sorts of different sources, but those sources are published. Ultimately it's a large language model, so they must be already in language land. They're in text and numbers. What Tree Ferrers Agents is doing, it's going off and making those observations or gathering the results of previous observations from satellite and scientifically robust models. So it's sourcing its information from a completely different domain, a different realm. So we're surfacing now that internally I'm very excited already about the results. It's telling me answers to questions that I hadn't even thought about, uh, as potential effects of things, El Nino and other things. So we'll be exposing that to you know, kind of select client, uh, in the next months. Uh, and I'm really excited about people being able to use that. The important thing for us, of course though that uh, agent is only valuable if it's surfacing things that can be relied upon. And so that's our, you know, our brand promise if you like. We've got to make sure that that data is not only discoverable and understandable, that it is trustworthy.

Speaker A: That sounds like uh, a little bit of a fresh perspective of how to build an agent. Like you said, it's, it's not.

Speaker B: Yeah, it's different. It's a different way of. Yeah.

Speaker A: Jonathan, as a closing question and um, if I would have to force you to give a piece of advice to supply chain or procurement executives now and uh, in terms of how they should start thinking or acting differently, uh, if they want to be prepared for the next wave of risk and disruption, whichever it might be, what would you, would you say?

Speaker B: I never give advice because I'm not very good at it. So uh, but what I would say is, what I reflect on quite a lot is I think about, and Covid's a great example. What happened in the first three, six months of COVID was quite phenomenal and all of us were kind of, in one way or another, really surprised by certain outcomes in positive and negative ways. So I would uh, ask people to say, you know, it's a fact of physics that things are going to get crazier that will absolutely impact supply chains. No supply chain will be immune from these effects. So I just ask people to say, think about two or three examples of during COVID either positive or negative, where you were really wrong footed, something that happened from one day to the next. And just imagine what that would be like if that happened in your, you know, sourcing of some of your key ingredients. And I'm not talking just like catastrophizing stuff. I'm talking about what if there was a prolonged shift in the source of cocoa and where this could really come from? What would you actually do? And because we always tend to as human beings think, oh, it'll come back. Oh, it'll be fine. It won't be like this for long. But what if it is going to be like that for time? A long, a long time. And so I think if we spend a few minutes thinking about, you know, what if it did really happen. I think that's a really useful mental exercise.

Speaker A: On that note, Jonathan, I want to thank you very much for joining us. Um, I think this was quite eye opening. You know, I have a lot of these type of conversations, but not from this perspective. Okay. Think about what could get quite disruptive because we might be there for a long time.

Speaker B: M. Yes. And Easter eggs getting smaller will be the least of our worries.

Speaker A: Good point, good point. Thank you again. All the best to you and the trifera you thank and we definitely stay in touch.

Speaker B: Pleasure. Thank you very much and thanks for the opportunity. I really appreciate it.

Speaker A: Thank you for listening to our podcast. For all the show notes and information discussed in the episode, please follow elcatglobal.com podcast also, if you found this interesting, please subscribe to the podcast on itunes, Spotify or one of the podcast platforms. We are looking forward to your feedback.

Speaker B: Sam.