[00:00:00] Speaker A: We run a diligence process that involves understanding technology itself. What's different? Why now the industry? Is this a compelling commercial opportunity? Great technology opportunities and great commercial opportunities don't always align. And finally, is this a team that we see as credible in pursuing both that technology and that market vision? Assuming we get comfortable with all of those things, we make an investment.
And then the question is, okay, how do we do as much as we can as advisors, potentially as board members, to de risk this business to make it more likely that they're able to attract capital?
[00:00:49] Speaker B: Hello, everyone, and welcome to the Robot Industry podcast. My name is Jim Beretta and I am your host. And it is my pleasure today to have Caleb Appleton from Bison Ventures on the program. And he's one of these people that actually fund robot companies.
And Caleb, welcome to the podcast.
[00:01:07] Speaker A: Yeah, of course, it's great to be here.
[00:01:09] Speaker B: And you're with a company called Bison Ventures. And can you tell us a little bit about Bison?
[00:01:14] Speaker A: Yeah, I'm happy to. So Bison Ventures is a venture capital Firm Founded in 2022 by my partners Ben and Tom. I joined them about a year later in 2023. And what we do is we back early stage technology businesses in a category we call deeper frontier tech. What that means is that every business we back and invest in has true kind of science or technology innovation at its core. We're not backing, for the most part classical enterprise SaaS, businesses where most of venture capital spends their time. We're, we're really focused on businesses that we think can drive step changes in industries, you know, move 10x better, faster, cheaper, and a lot of that ends up being in the physical world. And so there's kind of three categories we spend our time in, primarily the first of which I spend very little of my time, but the rest of the team has quite a bit of coverage on which is climate and sustainability. Though there are some applications of robotics certainly within that space. The second bucket is an area we call tech Bio, but is the intersection of technology toolkits and engineering toolkits with the life sciences. A space actually where robotics and automation have driven vast kind of improvements in throughput and learning cycles.
Then I spend the bulk of my remaining time in our final bucket, which we call AI in the physical world.
This is that interface of artificial intelligence systems with not just bits of information, but actually the atomic world. And that includes everything from true physical world manifestations of AI, things like robotics and automation, where you be truly moving something around using artificial intelligence through to software toolkits. That power those industries and those types of technologies.
We're an early stage investor through and through. And so what that means is we're investing at earliest kind of at this stage of idea, the kind of canonical two individuals with a pitch deck and a garage that they've rented through to the series A and sometimes a series B time point and the company's life cycle. And so inception moment is usually called the precede or seed moment, the initial seed of capital into that business that allows it to grow. The Series A, we tend to come in a little bit later. Maybe Those companies are 18 to 24 months old, proven out some of what was on that piece of paper. And maybe they've moved from a garage to, you know, warehouse space.
And then Series B tends to be kind of commercial stage businesses where we're really less financing technology risk and much more thinking about the market and growth.
[00:04:03] Speaker B: That's great. Thank you very much for explaining that, because I think it's always a little bit confusing about what is a series A and B. And so we're glad we've got that out of the way. So tell me about Caleb. Like, how did you get into this business of funding robotics and AI?
[00:04:15] Speaker A: I grew up loving science and technology.
I went to school and I studied biomedical engineering.
I was very interested in how you could use emerging toolkits in biology to engineer the world around us, the living world, including our own cells. I had the privilege and honor of working in a lab for one of the first labs in the world building on a technology called CRISPR for the gene therapy space back in 2012. But I came to realize that I didn't really love research and I didn't want to get a PhD. It just wasn't how I saw myself spending my time and spending it enjoyably.
I did general consulting for large multinational organizations primarily, and through that work got exposed to a number of kind of critical questions at the highest levels of those companies. But I still yearned for being kind of closer to the bleeding edge. And by happenstance, a recruiter reached out to me about a role at a different venture fund called Innovation Endeavors here in the Silicon Valley about a decade ago. Now, a lot of what we were then looking at was the earliest manifestations of what now has become commonplace as AI. At the time, people kind of said we were crazy. We were backing robotics businesses. We were backing applied AI businesses at the time. This is the pre transformer. AI was a very different thing. It was largely computer vision.
And what we found relevant to this discussion is it Was really, really hard to build those businesses. They required a lot of time, a lot of capital to build.
As an example, we had a company in the agriculture space building agricultural robotics.
They had to build the actual robot, you know, the hardware that was going on the back of a tractor. Not the friendliest of environments. Two, they were using computer vision to identify weeds and other things growing in these row crops. And so to do that, they had to build a massive data set.
You think of that TV show, Silicon Valley. Hot dog, not hot dog. It was that. It was weed, not weed. The company had to spend a lot of money building that data set, building that hardware, doing all of these things. And that company ended up being successful, was acquired by John Deere. It took much, much longer to build that business than I think it would today. And so I actually stopped investing in robotics for a while. For that reason. I said, this is too hard. And about two, two and a half years ago, I started to question that hypothesis again and have now made several robotics investments in the latest wave of businesses. Because I do believe the challenges faced by robotics businesses are becoming more tractable than ever.
And what it looks like to build a robotics business in 2025, soon to be 2026, is very different than what it was in 2017 or so.
[00:07:17] Speaker B: So did we answer the question about what's unique about the Bison team?
[00:07:21] Speaker A: We did not.
Let's do that. Happy to answer that. I think maybe it's helpful to give a bit of an oral history on venture capital in general and why my partner Sven and Tom ultimately founded the firm and why I joined them in that journey.
Venture capital. The first ever venture investment was into a computer company in the mid-1950s. That business called the Digital Equipment Corporation, resulted in a something like 150% IRR, very successful IPO and really spurned the creation of the venture capital industry. That industry for the first several decades of its lifetime. Founding technology businesses, things that were incredibly risky, largely silicon based computing businesses. That's why the Silicon Valley had its name. Yes, in the kind of 90s and early 2000s, those types of businesses fell out of favor with venture because they were highly risky, they were highly capital intensive. And at the same time the Internet was becoming a thing, the consumer Internet, followed by the transition to cloud and to Enterprise SaaS. And all of a sudden you had these businesses that were maybe expensive to get started, but very quickly could sell a product that was ultra high gross margin. And you could look at a couple of metrics.
What is the lifetime value of a customer, how much does it cost to acquire those customers? What's my retention rate? And you could make the job much more formulaic. And it became less about technology hunting and diligence, and more about making sure that you could spot the winners faster than other people. And so we at Bison felt that that left a huge gap in the industry. These transformational businesses to be built on breakthrough technology.
And each of us has spent our careers in venture backing businesses like this. And so Bison was ultimately founded to a dedicated platform with a team of professionals who have spent their kind of investing career focused just on highly technical businesses.
Because we believe that entrepreneurs in those spaces are going to be successful. It's very, very helpful to have not just the capital from people like us, but the experience and the guidance and the advice that we can bring to the table. Because we've seen this over a decade and across many companies collectively.
And so that's really how we differentiate ourselves is my money's no greener than the next guys.
It's really about the fact that we spend our time totally focused on just these types of businesses.
[00:10:10] Speaker B: That's. Thank you for that. And this is a good segue into my next question about how do you partner with a Robotics founder? If two of us come to you and say, hey, listen, we've got this great idea. What's next?
[00:10:21] Speaker A: You know, Bison has to decide whether or not this is a good fit for us from an investing perspective. And so we run a diligence process that involves understanding technology itself. What's different? Why now the industry? You know, is this a compelling commercial opportunity?
Great technology opportunities and great commercial opportunities don't always align. And finally, is this a team that we see as credible in pursuing both that technology and that market vision? Assuming we get comfortable with all of those things, we make an investment and then the question is, okay, how do we do as much as we can as advisors, potentially as board members, de risk this business to make it more likely that they're able to attract capital downstream from us, that they're able to bring a product to market that resonates and solves a problem.
And so at the very basic level, joining the boards of most of these companies, we're spending active time with these entrepreneurs. With every entrepreneur where I'm on the board of their company, I'm spending, I don't know, somewhere one to two hours a week on average with them focused on the most critical issues currently facing them.
So we tend to walk through their top priorities. We spend time making sure those are the right Priorities. And then we spend our time diving deep into the things that are stumping them.
It's a lonely job to be the CEO of any company.
Definitionally, you are the only person in that role. You do not have peers.
And so I can help to play, to provide guidance, feedback, and a sounding board that just does not exist within the walls of the company that they are building. In one week, that might look like, hey, I think I need to recruit someone that does X. What does that role look like? Can I actually get them? How would I go about finding them? Or two, I'm having this problem with this early customer. How might we work through that? Or it might be internal personnel related. Hey, my team is not working as effectively as I think they should be. What are some strategies we might work on to, to help overcome that? And so it evolves over time, but it's really about being that kind of partner to the entrepreneur that they don't have in their business.
[00:12:43] Speaker B: You're after the big Caleb. You're after the big successes for your investments in time. Like you're after hockey stick growth. Like, not, not, hey, we think we'll sell a thousand. You want to sell like 100,000, right? What does that look like?
[00:12:56] Speaker A: Yeah, absolutely. The way to your earlier question, you know, many of your listeners might not be as familiar with venture capital.
Venture capital is very much not. Like, maybe, you know, the money you have in your retirement fund where you're very focused on generating kind of repeatable and reliable returns without much risk.
Venture is really there to create outsized returns by taking risk. And so what that means is in our first fund, it's $135 million fund, we'll invest in about 50, 15, 15 companies.
And we expect that only a handful of those two to three will drive most of the returns of the fundamental. Because of that, you have this kind of power law effect where really the fund math only works if you have one to two companies. That $5 million investment returns that whole $135 million of the fund. And to do that for owning 10 to 20% of these companies, VR investment, you need to create companies that are able to exit for multi billion dollars.
And so to do that, it means they must be attacking massive markets and selling lots of widgets into those markets, whatever that kind of widget may be. And so we spend a lot of time doing a couple of things. One is when you're taking more technical risks in the market broadly, you don't want to then stack upon that unknown market risk. We Spend a lot of time upfront trying to understand if we build this thing, whatever that thing may be, is there a willing and excited buyer at the other end of that journey?
And the good thing is, for many of these companies, that's a knowable question.
We very much avoid the if we build it, they will come type of stories.
That's a good way to end up with something really exciting and expensive that no one wants to buy.
And two, we spend a lot of time thinking about what does this technology do to accelerate the industry and does that as a result cause it to grow more quickly? Or is the market size today a good predicate for what it might be once this technology is released? Really with that focus of, hey, to drive those multibillion dollar outcomes, you have to be able to deliver hundreds of millions of dollars of revenue, which means you very few things in the world deliver hundreds of millions of dollars of revenue unless they're critically important to some large industry.
[00:15:40] Speaker B: What are some of the red flags when you're looking at an opportunity or looking at entrepreneurs?
[00:15:45] Speaker A: Yeah, absolutely. Maybe on the opportunity side.
First, I think one is there's not a well defined, you can't really be.
You can't exit a conversation scratching your head about why this matters.
You have to. It should be abundantly clear that, okay, I don't know if they can solve this problem, but if they do, it's so obvious why it needs to exist.
The second thing when we look at the opportunity is, particularly in things like robotics, is what do we need to believe from a technology perspective for this to ultimately work. We tried to avoid being in the land of hypothesis risk, where there are these binary card flips along the company's development life cycle, which is like, will the technology work or not? Typically, we're investing much more at what we would call the engineering risk side of the spectrum, which means maybe all of the individual components have been proven and proven to work reliably, but they've never been integrated into a single system or there's some amount of scale up that needs to be done to make the economics work. But fundamentally you're not playing science, you're playing engineering as it relates to teams.
Fundamentally, we invest in people as much as we invest in companies.
And probably the lesson I've learned most over my career as a venture investor, and I continue to learn this lesson today, it's, it's a class that never ends, is that it is much, much a much better strategy to invest in exceptional teams where you might have some skepticism around the market or the technology, because those exceptional teams often find their way to overcome that. It might look like the business pivots or matures in a slightly different way or the market opportunity shifts slightly than it is to invest in technologies or markets that I feel rock solid on.
But I'm so, so on the team.
It is very rarely looking back at my portfolio and my partner's portfolios that the companies that don't work die because of technology impediments. It's much more because the team couldn't sell the product, they couldn't understand their customer need deeply enough, they couldn't iterate through adversity.
And so we spend a lot of time, I spend a lot of time getting to know the teams before I invest them both, asking them deep questions about their business, but then also trying to, to get to know them, you know, outside of just the pure, you know, four walls of the business that they're building and trying to assess that tenacity that ultimately I think is so critical for a founder to be successful.
[00:18:55] Speaker B: Caleb, what themes in automation and robotics are you most excited to back right now and why?
[00:19:01] Speaker A: You know, we're seeing a ton of attention poured onto the robotics industry from the venture capital asset class writ large.
People are looking for a similar breakthrough moment that we saw with ChatGPT in 2022 in the physical world with robotics. And I think my belief is that that moment is certainly coming and we're going to see really kind of unimaginable advances in both what the hardware can do and the tasks that are tractable. I think the question is on what time scale and what tasks and challenges come first. We're very excited about a couple of different themes. I think that the first is focused on the dull, dirty and dangerous, the jobs and tasks that are putting humans in harm's way and asking them to do things that are not well aligned with their own health and safety, that are very tractable for robotics and increasingly so as you have robotics that are much more able to respond to non deterministic situations.
And so that might look like we have a company called Bonsai Robotics, which both works in the ag autonomy space, but also has early kind of work in other areas like mining, you know, really focus on just outdoor autonomy broadly. And so that's one, you know, where is it that there are large amounts of people doing work that is dangerous and highly demanding, that we can use robots and cobots to make that work better and free up those resources to do much more meaningful and impactful jobs? We also have A company called Cobot in the space that's in the kind of warehouse, not just warehouse logistics, but kind of material movement space broadly across everything from warehousing through to events and hospitals.
Much of what makes the world go round is moving things from point A to point B. And Cobot has built a really flexible platform that makes that easier for operators of large facilities, regardless of their kind of specific industry.
So that's one I think the second area that I'm very excited about and interested in is companies that through the deployment of their robotic solution are able to capture rich, interesting and in context data.
If you go to any robotics conference today, particularly those that are more technical and focused on AI, the thing that is a constraint for everyone is the data on which they are training their models and fine tuning them against specific tasks.
I think that the most interesting companies are those that have an opportunity to one deliver something commercially useful today that isn't built on a requirement of having that data, but in doing so are able to create a rich data set that's that potentially very useful for themselves in building the next iteration of their product, but then also in kind of powering the field more broadly and then the final bucket, and this is one that I'm spending more time in, but haven't yet made an investment in, is looking at kind of horizontal tools that work for all robotics companies and that might be at the model layer, companies like PI and Skilled and Generalist and others that are building foundational models for physical intelligence. Or it might be at other layers in the stack where it's hey, I need a solution for sensor fusion or for safety or many other perception and planning.
The reason why One of the reasons why the kind of SaaS craze took over venture capital over the last two decades was because it became increasingly easy to build these businesses because entrepreneurs had all of these tools they could draw from.
So that if I wanted to create a new HR product, I didn't have to start from ground zero. Each time a new company was started. I could use all of these kind of composable tools so that I could focus just on the thing that I did really, really well. Whatever that kind of HR tool was, I was building.
I see the future in robotics being very similar where we kind of move away from every company having to build it all themselves to being able to use many high quality tools so that the company can focus really just on that last mile problem, whether it's the manipulation, whether it's how they navigate in a harsh environment, if it's the hardware, if it's the model, but not having to build it all from scratch every single time. The challenge I'm running into in that space right now is we're in the early days, we're in the Internet in the late 90s where we don't yet have massive companies built in, in robotics or that we have only a handful of them. And so there's only so many customers that can buy that product. I think that these will be, that that will be changing in the coming years and I'm excited to see how some of these more kind of horizontal tools end up growing alongside the industry.
[00:24:30] Speaker B: Thank you for that. And my next question is kind of a, is kind of a bit about robotics. Like robotics is hard because it's both hardware and software, AI and marketing, and long leads to development.
I keep thinking, could you have picked an easier sector? Like how, how do you compress this time cycle? And I get the fact that hey, I'm going to buy the arm from here and I'll buy the wrist from here and, and such, but how do you compress it?
[00:24:55] Speaker A: Yeah, it's a good question. We started the conversation talking about what it was like to build a robotics business, know, a decade ago and the time cycle was massively long. There were not good supply chains for really anything. You know, UR and Kuka and ABB were around, but they were incredibly expensive.
They did not allow much aftermarket kind of modification to occur on those platforms. So the vast majority of new robotics businesses built it all from the ground up. And that meant maybe years of development.
And when you're only building a dozen of your thing at a time, it's really hard to jump to the front of the line at a manufacturing partner or to get a company to pay attention and do something unique for you. And so I think what's changed over the last decade is we've seen many commercial off the shelf parts necessarily specifically focused on robotics come to play, crucial parts of the supply chain and robotics. And so that's everything from the sensor stack, I think lidar and onboard radar and other things are a fraction of the price that they were a decade ago, driven largely by the autonomy sector. And in vehicles pushing those costs down. The second is onboard power is improving drastically.
We have things like scooter batteries and wheelchair batteries that are very good that are in a form factor that makes sense for robotics that cost significantly less. And then again, you can buy you used to be the cheapest 6 degree of Freedom Arm you could buy was 50 to $70,000, US now you can order.
There's development kits that cost a couple hundred. There's ones that you would be comfortable putting in an industrial facility that cost a couple thousand.
That's all to say, I think the hardware ecosystem is maturing, and the best teams are no longer spending their time fretting as much about that. They're actually figuring out how to be really smart integrators of many existing technologies and then innovating as thinly as possible on the hardware side of things, where it really matters. That might be the end effector you're using. It might be, you know, something related to the joints. It depends on the company. But the less you can do, there is often more.
Yes, the second piece is, yeah, absolutely.
These companies are multidisciplinary in nature. And what it takes to be successful requires not just hardware talent, but incredible software and controls talent as well as, you know, the business talent around the table to make sure that you're really building something people want to buy and that, you know how to sell it.
At Bison, we love these interdisciplinary projects. We think that this is where you see synergy, where one plus one equals, you know, far greater than.
And the real key is making sure that you're able to build teams and structure businesses in a way where that can be true and that those aren't, you know, kind of siloed work streams. And so that is all to say, it still remains hard. I think that we are seeing a rapid pace of innovation across each of these domains that makes it faster than it's ever been. But it's never going to be software, particularly software in the age of AI.
But because of that, we think these businesses are more defensible, that they will have longer staying power and ultimately larger kind of financial outcomes for the funders as well as the founders of those businesses.
[00:28:44] Speaker B: I was going to ask about the core values and value proposition. We kind of reached those to being successful in the sector. How do you coach entrepreneurs not to aim for the moon?
[00:28:55] Speaker A: Yeah, it's a. It's a good question.
I might turn it around and say I. I do coach them to aim for the moon, or maybe beyond the moon. You know, the biggest companies that have ever been built have been people that have been built by founders and teams that dared to dream bigger. I think one of the reasons Silicon Valley has been so successful is that venture capital ecosystem really is predicated on this power law returns math. And so people are much more focused with the home runs than they are the base hits. And that pushes teams to think bigger. That doesn't mean that people can suspend reality and operate and you know, a different non Newtonian physics regime than the rest of us. And so the balance is always how do you cast a big vision and have a clear line of sight to how you unlock that vision, but then also ensure you're being pragmatic and that there are validation steps along the way that confirm that big vision is realizable. And so what I coach founders not to do is have these grand reveal kind of waterfall style development plans.
We believe truly at bison that the best deep tech businesses de risk not just tech along the way, but you know, commercial traction along the way. And so we spend a lot of time pushing our founders to ensure that they are speaking to customers, that they're putting product in the hands of customers as early as possible.
And I think that makes that big vision much more kind of compelling and realistic to people downstream of us, the next set of investors who might come in into this business. And so I really do my best not to push founders away from dreaming big because I think that is what drives large amounts of innovation.
But then on the day to day, I make sure that we're doing everything in our power to ensure that we are eliminating as much risk from this business as possible with each dollar spent.
[00:31:15] Speaker B: Yeah, and I guess that answers the question about making a robot company scalable or suitable for commercialization, correct?
[00:31:22] Speaker A: It does. And I think this is again something that's changed over the last 10 years is it used to be you would spend three to five years of development to build something, only to find out, hey, when I deploy this, it doesn't work well enough for it to actually be useful for my customer. That's a really hard way to build a business. Now we're seeing companies that within 6, 12, 18 months, you know, after first dollars into the business, they have robots and devices deployed at, you know, pilot customers where they're able to get this real and extremely valuable feedback that allows them to iterate more quickly and ensure they're building something that solves the industry's problem.
And that's, you know, fundamentally where they have to focus. You know, that's why mobile apps were so great is you know, they launched them in five days and people use them or they didn't and you make a change and people use them more, they use them less.
We're never going to get to that level of iteration cycle in robotics, but I think we're to a place where you're not just speaking in the abstract and in the hypothetical, but you can put something in front of a customer and you can learn from how they use it, not just from your hypothesis on how they might use it.
[00:32:41] Speaker B: Caleb, that was great. Did we forget to talk about anything today?
[00:32:46] Speaker A: I don't think so. I think it's been a really exciting and interesting conversation.
Speed of progress across the industry over the last three years, I think has. If you'd asked me three years ago where things would be today, I think I would have undersold probably the progress that has been made. And I expect that experience to continue to be true for myself, where I'm pleasantly surprised to the upside about how quickly the technology in the industry is moving.
[00:33:15] Speaker B: So when you're not helping build robot and AI companies, what do you like to do? Do you have any hobbies?
[00:33:20] Speaker A: I do, very fortunately, I have.
For me, this job is all about balance. And I think I'm more effective at the job because I'm able to find balance out outside of it. And so the thing that I do to drive clarity in my thinking and reset myself and ensure it sustains sustainable for me is I'm very passionate about exercise and fitness. The things that I'm very excited and interested in is endurance athletics primarily. And so I live in San Francisco.
Most mornings you'll find me on a mountain bike, a road bike with my trail running shoes on, across the Golden Gate Bridge and the mountains of Marin, you know, enjoying a sunrise and being able to be fully disconnected from technology.
And I find that that's both obviously good for my physical health, but mostly good for my kind of mental health and makes me much more kind of clear and focused throughout the day as well.
[00:34:29] Speaker B: Yes, you certainly live in the beautiful area of the world and how could people get a hold of you? If they want to find out more about Bison and about Caleb, feel free.
[00:34:37] Speaker A: To reach out to me directly through LinkedIn.
That's probably the easiest way. And Bison www.bison.vc and you can find out a little bit more about our portfolio, the others on our team. And I'm always excited to chat with people who are starting a business who might be thinking about starting a business.
I'm really, you know, I've many people have given me their time over my career. I'm very happy to kind of pay that forward.
[00:35:09] Speaker B: Thank you for joining us today, Caleb. I look forward to listening to the podcast.
[00:35:14] Speaker A: Yeah, likewise. Thanks, Jim.
[00:35:16] Speaker B: Thanks everyone for listening. As you know, as you might know, I run a marketing consultancy called Customer Traction where we focus on marketing, branding, strategy and content. Creation. We do a lot of project work and help you with your marketing challenges, and we're focused on automation integration and robotics. We welcome new customers and projects who can help you get your marketing back on track, fix your website or perform a marketing audit. And thanks to our sponsor, Mecademic Industrial Robots. Mecademic builds the world's most compact and precise robots used in industries such as photonics, medical device, optics and electronics. Mecademic continues to set new standards in precision, footprint and flexibility, accelerating small component automation for manufacturers with robot integrators and innovators alike. You can find
[email protected] and I'd like to acknowledge a three the association for Advancing Automation. They are the leading automation trade association for robotics, vision and imaging, motion control and motors and the industrial artificial intelligence technologies. Visit automate.org to learn more. And if you'd like to get in touch with us at the Robot Industry podcast, you can find me Jim Beretta on LinkedIn.
Today's podcast was produced by Customer Attraction Industrial Marketing, and I'd like to thank my team Chris Gray for the music, Jeffrey Bremner for audio production, my business partner Janet and our sponsor Mecademic Industrial Robotics.
