.png)
Another ClimateTech Podcast
Interviews by Ryan Grant Little, a climatetech founder and investor that explore the fight against climate change through with founders, investors, activists, academics, artists, and more.
#Climate #Climatetech #Cleantech #Sustainability #Environment
Another ClimateTech Podcast
Cutting Heavy Industry Carbon through AI, with Leise Sandeman of Pathways
Leise Sandeman is co-founder of Pathways, a company using AI to make heavy manufacturing more sustainable. Based in Copenhagen, she brings deep expertise in steel production and AI to the challenge of industrial decarbonization.
In this episode we talked about:
🏭 How Pathways is building the data layer for sustainable manufacturing, with their software now deployed across hundreds of plants
⚡ The current reality where sustainability teams spend 90% of their time collecting data and only 10% implementing changes
🤝 The "playground" mindset of being a first-time founder and the importance of co-founder trust
🌍 Taking a strong stance on being a "single planet society" while working with diverse perspectives across regions like Texas
#ClimateTech #manufacturing #sustainability #EPD #LCA
🪸 Transform your company's milestones into impact, like trees planted and coral reef restored: impacthero.com/podcast
🧑💼 Growing across Europe? Grab a free consultation and hire without hassle: parakar.eu/climate
When people in the climate space talk about this amorphous scope three, it's something they assume and top down. So going down into that supply chain and really developing software and solutions that work for manufacturers was how we went and came up with Pathways.
Ryan Grant Little:Welcome to Another Climate Tech Podcast interviews with the people trying to save us from ourselves. Leise Sandeman is co-founder of Pathways, a company using data to make heavy manufacturing more sustainable. She's passionate about things as disparate as steel production, ai and biodiversity, which is probably a great combination of interests for a climate tech founder. I reached Leise in Copenhagen. I'm Ryan Grant Little. Thanks for being here, Leise. Welcome to the podcast.
Leise Sandeman:Thank you so much for having me.
Ryan Grant Little:You're the co-founder of a company called Pathways, which aims to decarbonize the manufacturing process, and right now at least, you're focused on building materials. Where did the idea?
Leise Sandeman:for Pathways come from. I always love that question because, you know, there's this conception that you wake up one morning and you've had this brilliant idea for a solution that everyone urgently needs. Me and my co-founder, alex, likes to say that the idea from Pathways really came from our customers, but there's, of course, more pieces to it, so a few different pieces of the puzzle here. We'd spend a lot of time speaking with real estate developers, architects and general contractors and they kept saying we're missing data. We can't figure out the data of the buildings when we build them and how to decarbonize them. And I'm simple and I'm from Denmark, so I think of buildings like Lego. Right, they're the sum of the bricks.
Leise Sandeman:And when you went down and looked at the bricks, it became really clear that manufacturers were struggling to deliver the product carbon footprint or, in the kind of terminology, environmental product declarations and lifecycle assessments and that pressure and demand was really it. The second part was I spent a while working out in India with Tata Steel and you know, you stand in Jhansshapur and all you can see is steel. The size and gravity of the challenge became really clear. But what also became clear was that manufacturers were largely left alone without any tools or software and help on how to actually understand their own supply chain. They're really everyone else's scope three, but when people in the climate space talk about this amorphous scope three, it's something they assume and top down. So going down into that supply chain and really developing software and solutions that work for manufacturers was how we went and came up with Pathways.
Ryan Grant Little:I love that because, as an investor myself, I have a bias towards companies that work with their customers first and say there's really a problem here and they build a solution, a product to serve those customers and solve that problem, Versus the other more top-down approach of I have a really cool idea. Let's just build this thing and see if anyone wants it. The field of dreams if you build it, they will come. That often doesn't work in entrepreneurship, but finding something where you're embedded in the space and like, as you say, you know steel's a very good example and you just you see this and you understand kind of the wickedness of the problem in the view of the companies that you know that have it, and then you come up with ideas for how to solve that.
Leise Sandeman:No, you're absolutely right, and the same goes for I mean, it's a continuous journey. You don't just come up with your product right. It's investors talk about the infamous product market fit right, do you have the right product? Does it fit what your customers ultimately want and is there scale in it? And we talk a lot about that as kind of you have your hand on something that's hot and you've got to continue following that right. So as soon as you have a few customers that are asking for something else testing and developing that, is that working? How can that deliver value? It's really a continuous journey, much more than a. You woke up one day and you had an idea, and then it was done.
Ryan Grant Little:Yeah, product market fit is a really moving target, right. I mean, the problem changes. The features that you came out with five years ago are commoditized now and you have to kind of keep ahead of the curve it's like going to the gym.
Ryan Grant Little:It's an everyday activity well, I don't know if I, I mean, I agree, I agree with that in principle, but set some of the numbers here. So what do we need to know, kind of about the environmental impact of construction materials? And I mean you mentioned steel, which I would imagine has a massive footprint.
Leise Sandeman:Yeah, let's start with what we'd call the built environment. Right? This is the most shocking fact, in my mind. The next 30 years, we're going to build more globally than we ever have in human history. Every month, we're building the equivalent of a new New York City. Americans love that fact, and everyone deserves good and safe housing.
Leise Sandeman:It goes without saying.
Leise Sandeman:Right, and particularly as we're seeing rapid climate change, adaptation and need for protection and changes in how we build our infrastructure is also increasing, and all that is corresponding to a huge demand on building materials.
Leise Sandeman:So globally, right now, the built environment is responsible for 40% of emissions and a lot of attention has been put on heating and electrification but 11% globally of emissions. That's just greenhouse gas. That's not even talking about waste and biodiversity. It's just pure materials, and this bucket is looking to increase rather than decrease, and right now we have very few solutions, and so you know, when we think about the numbers here, this is an area that's not going to be helped very much by electrifications and the innovations we're seeing in the energy space. It's also incredibly dispersed globally, which means we need there's very few silver bullets. I like to say it's a game of slivers of silver, and that's really why I mean we think it's so important to sit with every single product and improve, whether we're working with a customer that has an innovative product and they want to show how much better it is, or with an established corporate that's looking to reduce one or 2%, because that really shifts the needle.
Ryan Grant Little:So, and who are those customers? Maybe you can walk us through a little bit kind of what the platform looks like. And I mean, are the customers both these corporates that are looking to reduce their footprint and the manufacturers of materials? Is it in that sense kind of like a platform or do you, is one or the other more of the customer?
Leise Sandeman:Yeah, so I'm going to ask anyone listening to this podcast right now to kind of close their eyes and stick out their head.
Ryan Grant Little:Unless they're driving.
Leise Sandeman:Unless they're driving. If they're driving, you're not allowed to close your eyes In which, in which fact, actually keep your eyes open, but look around right, the window where you're sitting, the steel beam in the house you're in, the bricks, the carpet, anyone who produces any of those materials. That's our potential customer. So our customers are companies that have production sites, have plants, who actually make something. That's a whole host of different types of companies, spanning from big companies that make cement and ready-mix concrete, steel manufacturers, utility poles, complex wood panels, carpets, down to kind of what I'll call less of our core customer type. But customers we like to work with which are other innovative companies that are making innovations in the space. But the majority of our customers are large companies that are looking to understand the product footprint across a portfolio of maybe a thousand products or 10,000 different SKUs.
Ryan Grant Little:So let me see if I understand this correctly. So basically, you're collecting and processing data, you're generating lifecycle analyses, which are also known as LCAs, for the materials from this data, and then you're kind of condensing that into something called an environmental product declaration or EPD, which is a new term for me, actually and it's sort of like a verification document. That's the summary of the LCA, and then is this kind of the end product of what comes out of it. And yeah, answer that first, and then I've got some follow-up questions on that.
Leise Sandeman:Yeah, absolutely. Pathways sits as a platform that our customers get access to that shows close to real-time view of emission for each product. So we think of the lifecycle assessment like the engine. What do you need to create a lifecycle assessment of that steel beam that's in your house. You need to figure out, well, what was all of the input materials, how was it transported to the manufacturing site and what's everything that goes into the manufacturing process. In lifecycle assessment terminology A1, A2, A3.
Leise Sandeman:All that data sits in a host of different data systems out with our customers. The most typical ones we integrate with are ERP systems, procurement information, operating systems, and what we've become really good at at Pathways is basically pulling that data without our customers needing to do a lot of mastering or a lot of data handling, because we know how to transform those inputs into the lifecycle assessments. I had a call with a facility manager the other day where we were walking through our platform and as he was engaging and looking with it, he said, oh, this just looks like an overview of my production flow diagram and I said that's exactly what a lifecycle assessment is. It's an environmental digital mirror of your production process and that's really what our customers are able to then play with.
Ryan Grant Little:Digital twin. It's also known as right.
Leise Sandeman:Exactly. Digital twin is kind of like the amorphous term that's been going around for decades, right With the dream of how do we truly mirror the production process. But I think we're doing some of the closest that exist in the market to that which is being able to dive into a specific input of a product or a specific production process.
Ryan Grant Little:And I can imagine this works really well. But if the data is good, right, it's totally dependent on the data and then the kind of analysis and synthesis of that data into something that goes out. Where does the data come from? Is it self-reported or is it? How do you ensure that this is good data in?
Leise Sandeman:Yeah. So what was kind of our big unlock in a lot of this? Well, life cycle assessments in themselves are not that complex. They're certainly not rocket science. It's amount of something times intensity of it. What we found out is that, you know, finding the intensity factors of, for example, an electricity grid is not that complex.
Leise Sandeman:What's really hard for customers is actually finding their own data. How much electricity did we use in this production line, or how much water or how much waste was generated, or which supplier and how much are we using in each production line? But this is data that they keep right. They are either keeping it for production tracking, for efficiency reasons or for just kind of receivable rate on the procurement side. So our whole thesis is don't have customers put in data again or double down on data. This data exists in their systems and it's all about figuring out where it lives, building the API, roads, and then we can pull out that data. That doesn't mean it's perfect, right, we get some really, really messy data, but that's kind of you know, when we deploy AI on our side, it's a lot about structuring and sorting the messy data on our customer side.
Ryan Grant Little:Don't let perfect be the enemy of good and that kind of principle. I can imagine that a lot of the customers you're working with have been historically working with consultants to try to surface this kind of stuff and working kind of in a more manual process. And working kind of in a more manual process Is that kind of the process that you're automating and does it still require kind of a lot of you know, consultative handholding at the setup phase?
Leise Sandeman:Absolutely. I mean first of all. Yes, I mean, in some ways, pathways is a really simple business. We're taking something that was done all manually, paying very high prices for consultants externally for good reason, because they're spending tens of hundreds of hours on this, back and forth, sending excels. Can you fill out this document? Oh, you got the wrong data in here. Oh, this number doesn't make sense. I mean, the average wait time for working with a consultant to generate an environmental product declaration can easily be kind of 12, 18 months, mostly because you're going in and you're finding all this data manually and you're having to put it into a template and back and forth and building the LCA. So that is the process we are automating.
Leise Sandeman:What are the biggest differences when we think about onboarding a customer versus working with a consultant? Well, we think of it as it's all about data, which means you don't get around in some way, shape or form involving IT, and that introduces a setup phase, right? I was speaking with one of our lead engineers asking him well, how long does it take to build the API to this specific customer? He said, oh, you know, roughly two and a half minutes. And I said well, I can't tell that to the customer. You know it seems one completely unreasonable. The short answer here, though, is the technology part is not very complex. It's getting the right person from IT on a call, making sure they also have the context and having them give the API key so we can pull the data. The tech side of pulling data from our customer is not difficult, but there's always a people component.
Ryan Grant Little:You mentioned AI. You know I've been doing this podcast for a little over a year and at the beginning AI would kind of pop up occasionally, and now I feel like it's in every episode. It has to be because it's so core to a lot of the things that we're doing. I wonder if you could just talk a little bit about, I don't know your philosophy, your approach to using AI, maybe even some of the kind of nuts and bolts of it. Are you adapting large language models for your specific internal uses? In what ways are you using AI to build a bigger moat around you relative to competitors?
Leise Sandeman:I think the first thing here is our customers do not care if we use AI. They care about if we're solving their problem and from that perspective, ai is truly I mean, I believe it's truly transformative to how our whole industry is working, to how you build technology companies and how you scale solutions. But that's not how our customers are thinking about or why they're thinking about buying pathways, which means when we apply I mean we applied a few different places right now in kind of in our solution stack. The most important place is not Gen AI coming up with new data. It's training models on all the context of data that a customer might share.
Leise Sandeman:So we have a customer in French-speaking Canada. A lot of their fuel information are in French PDFs and LLMs are great at actually reading out all of that context and pulling out well, what was the fuel consumption use and summarizing that out. That's a fantastic use. It's kind of cutting down something that would have taken 20, 30 hours of manual work, now done in seconds. And the other part here is every manufacturer thinks of their process as really unique, but the data looks super similar. Everyone has input materials. Everyone uses a combination of electricity, fuels, lubricants, water, waste generation outputs in their process, which means we're able to kind of create quite a strong technology moat because we are passing through so much customer data from a variety of different manufacturers.
Ryan Grant Little:Yeah, that makes a lot of sense. I like your point also. You know this again comes back to your customer focus that they don't care how you get there, they care about solving the problem. I do a lot of pitch training, you know, for startups, and what happens a lot, especially with companies that are founded by engineers, is it's like, you know, here's one slide about the problem and solution, here's 10 slides under the hood, you know, showing all these things that we think are really cool. And then you know, maybe we'll remember to ask or to ask for money and tell you what we're raising for.
Ryan Grant Little:But actually, you know, as you say, and the analogy I use is like think of a car commercial, right, I mean, a car is a very, has a lot of technology in it, but how often do you actually see the hood pop open in a car? Commercial right? You see instead soccer balls coming out of the backseat and you know families driving around, because it's focused on the solution and the use case. And I think the same thing you know for any startup founders out there who are preparing to raise. This is good advice and, liza, that's probably served you well as you went to raise a million dollars in February 2024 and ended up raising 2.5 million, so a pretty decent oversubscribe rate there In the intervening year. From when we're talking now, needless to say, a lot's happened regarding politics domestically in the US, where you're largely based, and also sentiment, kind of on climate and climate tech and ESG especially. I wonder how have you experienced that change in sentiment, if at all, as someone who's leading a company like this?
Leise Sandeman:Yeah, I get this question a lot and in the week of the election I was actually down in Texas visiting a bunch of my customers in the ready-mix concrete space.
Ryan Grant Little:The Texan concrete space, famously blue.
Leise Sandeman:Blue state. Right, I can't vote in the US, but we probably wouldn't have voted the same, voted the same. Well, what does become very clear when you get out and you actually speak with plant manager, with facility managers, is a lot of them are taking initiatives, doing different things to improve either the kind of total emission footprint or the local impact of their factories and sites. Right, so most of our customers are not ivory tower climate defenders who are out talking about emissions. They're very real people who run plant operations, who have customers who are requesting materials from them Because, again, I mean, either things are getting refurbished or new construction or infrastructure is being updated, and I think a lot of our principle at Pathways is have respect for the fact that this is not going to be the number one concern for every single one of our customers or users, but it can still be a high priority and there are ways to make it much easier to engage with. So let me come with an example from our actual product.
Leise Sandeman:You were talking about the environmental product declarations earlier, right? I like to say they're like the environmental nutrition facts Instead of calories and carbohydrates, it's chemicals and carbon and, in addition to global warming potential, there's also a range of biodiversity factors, all with kind of ample Latin names that no one outside academia knows what means. Right, we've made a few very simple design choices in our product interface. So instead of eutrophication it says Algae Bloom. Right, that is the natural consequence of eutrophication potential of a product, but that suddenly means that someone who sits in a sales role for this kind of complex product feel comfortable speaking about it, because no one wants to feel stupid and in my experience, most people are willing to do a bit if you don't make it too hard. So again, it's probably back to the less of the silver bullet and more of the slivers of silver, but also a fundamental belief that people do want to make positive changes, and it's certainly something we're experiencing with our customer base when you make it much easier for them to see what are the changes they could make.
Ryan Grant Little:That's heartening to hear, and I mean my approach has been to kind of bake in the impact but not necessarily have that at the forefront of the sales pitch. So people generally want things to be easier, cheaper, more accurate. These are the types of benefits that people respond to, and if you can kind of pitch on the basis of that, but underlying that is greater efficiency and climate mitigation, then that tends to resound with people pretty well and I wonder if that's kind of the approach you're taking as well. I mean, presumably the work you're doing ultimately saves the customer a lot of money, headaches et cetera.
Leise Sandeman:Yeah, I mean at the core right, most of our customers used to have to spend, I mean, tens of hundreds of hours doing. I mean I cannot explain how boring it is to have to go into your own SAP system to pull out data, write it down on a piece of paper, to then put it over in Excel to then send it to a customer. It's not just boring, it's expensive and there's a huge labor shortage and there's not enough skilled people to do this. And then you have this booming ESG consultant group where you're just kind of pouring out money to people who also, fundamentally, would rather do strategy and real impact rather than measurement. I say you know, let's shift the.
Leise Sandeman:Today our customers are spending 90% of their time. I mean the customers and users who sit in sustainability are spending 90% of their time collecting data and 10% acting on it. I want to shift that around spend 10% collecting. We can do all that. We can help with the scenarios, we can help model all that, and then you can spend 90% on the change management that we so urgently need.
Ryan Grant Little:So, as we talked about, you've raised all this money last year. The company is growing. You're scaling up. What does the next couple of years look like in terms of company roadmap?
Leise Sandeman:Yeah, absolutely. I mean very exciting times for us. So I think of Pathways as building the data layer for sustainable manufacturing, right, this is an enormous, enormous market and we've just scratched the surface of it and been able to scale really fast. In that context, as many other kind of pre-seed raised, venture-backed startups, we're racing based on milestones and we're approaching those fast and certainly imagine that we're going to be going out and looking for more people to join our mission and more people to kind of join our cap table and help us run to those next milestones. I mean, in my mind, what's kind of success been looking like in the first year? Right, we went from having a pilot and no product live to now being live with 10 plus customers, you know, scaling out to hundreds of plants and that is incredible. But when we think about, well, growing to the next level, I mean we are our ambition is hundreds of manufacturers, thousands of plants and certainly also scaling outside North America, where we're already experiencing quite a lot of demand in Europe.
Ryan Grant Little:And you and your co-founder, alex Cooper, both have a lot of experience working in the startup world, but for both of you it's actually the first time your first time in a founder's seat and I wonder what observations you have about. You know this is still very fresh. It's been a very positive journey so far. What are kind of your observations on the experience of founding? What were some of the biggest surprises? You know for yourself and if you can speak on behalf of Alex?
Leise Sandeman:Yeah, I'll speak on behalf of Alex as well. We spent so much time together. I mean, I think the co-founder relationships probably intimate, intimate, close relationships you can have, right, you're building something together you are so passionate about that you're both invested in. And also, it has to rely on a hundred percent trust, right, the sensation that, as I turn my head this way and look at growing with our customers, he's building and scaling our product and our engineering org and, might I say, doing an incredible job at it. So, being a first-time founder, it feels a lot like playing when you were a kid. My description to my friends is like the sensation of being at the playground and coming up with a good game. And you need to. You know, you need to get your friends along, you need to get people to believe in it and it's all make-believe, right. So someone said, well, don't you feel imposter syndrome? And I was like well, no-transcript I've had previously required of me and it's a different space to sit in.
Ryan Grant Little:It makes me think of Yuval Noah Harari in Sapiens, talking about all these things that we hold as given that are all fictions like money companies. All these things are we hold as given that are all fictions like money companies. All these things are are fictions that only work if other people believe in them. Right? So I mean, and that's as you talked about the imposter syndrome and and being a founder, I mean, it really really seems to fit in that framework yeah, and then it does become real, right.
Leise Sandeman:Yeah, that's the you know and and I love this I mean, harari's concept of the story is at the core. But then you know, now, a year after we raised our pre-seed, we have a product that is ingesting real data, generating real emission footprints, allowing our customers to figure out where can they improve and ultimately also drive more sales based on sustainability. So this idea of kind of constantly coming up with the next story and the journey and the huge scale of the potential for what we're building, yeah, I mean again coming back to some of the pitch training work that I do and accelerator work.
Ryan Grant Little:Everything is about creating a story, making sure that all the important facts that need to be there are somehow inserted along the way. But you know, a 10 page, a 10 slide PowerPoint with eight numbers on each one is going to get you nowhere. So it's about telling the story of like, why are you the right person to do this, why is this important to you? What's your vision for this? And then along the way, sprinkling in kind of the important facts about market and customer and that type of thing along the way.
Leise Sandeman:Yeah, and I mean I always say for me it's all about our customers. They're the ones who are making the real impact here, and how do you put them at the center of the changes they can make in their own companies? And I mean I'll be forever grateful to some of our really early customers who took bets on us and kind of rolling out and testing what, what pathways could be, but ultimately in service of a mission they're on and a story they're telling inside their organizations about. Why should we do this with technology, with ai, instead of in kind of a more traditional manual way?
Ryan Grant Little:some of the advice I give to founders is that when you're talking with investors, you can never talk too much about your customer. There's no such thing right. I mean there's just. There's absolutely no. It's music to our ears, as investors, to hear that you're completely focused on your customers.
Leise Sandeman:Well, I'm almost obsessive, so that might it might tend to over done, but I don't think so. Catch me on a steel plant with a hard hat and I'll be hard pressed to leave.
Ryan Grant Little:One thing you posted on your LinkedIn which I love is you say that planet and climate is what keeps you up at night and that you live your life in pursuit of a single planet society, and it's the first time I've ever heard someone say that, and I think it bears calling out specifically now in this like Musk dominated, endless news cycle that we seem to be in, For you know this, we're living in tough times for to be optimistic about some of this stuff, and I wonder what advice do you have to listeners who also kind of stay awake at night worrying about this stuff and our only planet?
Leise Sandeman:Well, I think first of all believing in the single planet. It gets me into a lot of trouble Again sometimes when I'm down in Texas visiting customers. We certainly have different perspectives on that, but at the core I think we have a wonderful planet here and I want to dedicate my life to making sure it's accessible and available not just to humans but to animals, to our kind of wide range of biodiversity, for many, many hundreds of generations to come. The anxiety that I have good friends who feel family, who feel is very real, and I mean I think we all deal with that differently.
Leise Sandeman:I have a bias to action. It's probably why I ended up being a founder in the first place and for me there's nothing more. There's nothing that gives me more hope than the sensation of doing my part. But that can look really differently. Right Part of my job also includes that I fly much more than I would like to and it gives me a lot of guilt. But ultimately, you know that's part of being able to kind of build what I do and I certainly justify it for myself. And simultaneously, you know, you see people who handle and deal with that differently and see communities around how to do it. But for me personally, there's nothing more fulfilling than trying to do my part.
Ryan Grant Little:I think that's a great place to leave it and very good advice for people out there. That, yeah, I mean. I think the bias towards action is also a really good point, Leise, thank you so much. It's been such a pleasure talking to you.
Leise Sandeman:Such a pleasure. Have a lovely day.
Ryan Grant Little:You too. Thanks for listening to Another Climate Tech Podcast. It would mean a lot if you would subscribe, rate and share this podcast. Get in touch anytime with tips and guest recommendations at hello at climatetechpodcom. Find me, Ryan Grant Little, on LinkedIn. I'll be back with another episode next week. Bye for now.
People on this episode
Podcasts we love
Check out these other fine podcasts recommended by us, not an algorithm.
