🥗 Propagating the Indoor Farming Industry #015
feeding the planet via growing plants in boxes
TLDR: A synthesis of three interviews with leaders from three indoor farming startups, each with their own beliefs. A text-based arena where indoor farming ideas get battle-tested.
After exploring animal agriculture and regenerative agriculture, it felt natural to keep going down the food production rabbit-hole. In many ways, indoor farming (AKA vertical farming or controlled environment farming) is diametrically opposed to regenerative farming. The former recognizes the limitations of flat, traditional outdoor farming and addresses it through maximal control over the environment (temperature, humidity, water, light). The latter recognizes the inherent ecological value in how things used to be before the Industrial age of mass mono-crop farming. Although both employ modern day technology, there’s a fundamental difference in philosophy. One approach believes we solve the present-day problem by going back to the past (albeit with better tools). The other posits that we fix the problem that hyper-growth global capitalism caused by doubling down with even more modification to the natural world.
As kids, it was socially acceptable to ask anything. Our teachers and parents encouraged us with “There’s no such thing as a stupid question.” But as we got older and our curiosity waned while fear of judgement creeped in, we lost our ability to wonder. We’ve all at one point looked up at the sky and asked why the sky is blue. In exploring indoor farming, I embraced the naiveté of childhood and asked “Why can’t we just rotate land 90 degrees?” Land is expensive and flat. Why not go up? As we’ll see, there’s a lot more to indoor farming than just the dimensional transformation from 2D to 3D.
For this issue, I tapped three different indoor farming startup operators who spend every day thinking about the problem of healthy, sustainable food production at scale. I deliberately chose to include multiple perspectives over the typical one-on-one interview. To form my own climate beliefs and potential solutions, I’ve found it to be tremendously useful to absorb a range of opinions and then see what sticks. Climate is a nebulous, complex concept with far-reaching implications across every industry. Executing tangible, effective solutions requires rigorous thinking rather than betting that every approach is equally viable.
What follows is a synthesis of three interviews (conducted separately) with Mike Ross, Co-founder and CEO of Beanstalk (YC W18), Ryan Schools, Group Manager of Supply Chain at Plenty, and Jake Felser, CTO of Freight Farms. Rather than stack them one after another, I identified and interwove key themes together. It was insightful to learn that the three different approaches to indoor farming result in far less direct competition than I thought. Like many other areas of climate tech, the jury is still out for who comes out on top.
Introduction
Some quick context on the three companies:
Plenty is a vertically-integrated, vertical farming late-stage startup that’s raised over $900M in funding and has over 400 employees. They grow various lettuces and sell them to companies like Whole Foods, Bristol Farms, and Instacart.
Beanstalk grows lettuces in addition to other crops like herbs and microgreens. They take a slightly different approach to the larger, wholesale-facing vertical farming companies by using a soil-based approach (instead of hydroponics) and also sell direct-to-consumer.
Freight Farms doesn’t grow any produce themselves. Instead they create the indoor farm itself out of shipping containers and sell that (with additional products and services) to organizations and companies who will do the growing themselves. Their customer base ranges from non-profits to universities to five-star tropical island resorts (and more).
Topics
The Problem
Produce as the Product
Climate Mitigation vs. Climate Adaptation
The Role of Capital
Interconnectedness (Energy, HVAC)
Standardization
The Problem
What are the biggest problems with conventional agriculture?
Mike Ross from Beanstalk: Right now climate is the biggest issue. It's taking away farmland. We're losing millions of acres of farmland a year across the globe. It's also creating these unbelievable once in a hundred year type storms. We've seen that from both the mega drought that starved all of the crops of water.
We've seen crazy heat waves that come through and destroy an entire strawberry crop. And then now we've got the flooding and torrential downpour from the atmospheric river that is also wreaking havoc on farmland out west in the US.
Second to that is the seeds. We're using very specific seeds that have been selected for disease resistance, pest resistance, and for crops that are transportable. [Conventional farming] is not selecting crops that have nutrition or flavor. As a result, we've lost a lot of nutrition in our fresh foods.
The last one is the pollution of it. 70% of freshwater goes to agriculture and all of that runoff is full of pesticides and that's causing issues with our freshwater sources - so rivers, ponds, lakes. It creates these huge algae blooms and destroys ecosystems.
Plus the pesticides. It's a pretty rough environmental issue. There's very limited options in fighting any of those major issues. On top of all of that, there's a labor issue where immigrant workers aren't being paid enough. They're not being taken care of enough. And we still need more people.
So there's a huge lack of availability in the labor markets and the wages have to go up, which means food is gonna have to get more expensive. Those are the really major ones with climate absolutely number one.
…
We've discovered that the modern food system is almost entirely processed foods. In fact, roughly 60% of the average Americans’ calories come from ultra-processed food, which leads to a 41% obesity rate. In the US, it's just really, really difficult to eat fresh food.
It's so much of a special commodity that it's quite literally a specialty crop in the USDA's classification of food. Our mission is to build a new food system that’s based on fresh foods. To do that, we had to come up with a new method of farming that increases the taste, convenience, and sustainability of fresh produce.
Beanstalk is able to grow with 95% less water. Why is conventional agriculture so wasteful? And how are you so efficient?
Mike Ross from Beanstalk: Yeah, that's a really good question. So, On the traditional side you cannot give plants exactly the amount of water that they need because the weather is so variable. You're also not able to direct the water to plants that precisely. So there is a system of drip irrigation that's used for really valuable crops. Trees, vines, bushes, like anything that's gonna be multi-year. But as soon as you get into annuals, things that you have to replant every year the irrigation is much more wasteful. You also lose a lot to evaporation. And so you have to account for that with a hydroponic system or something like what we're using.
We know exactly to the milliliter how much a plant needs because we're controlling the environment it's in. And so we just give it exactly that much. Plus a very small buffer that can be recirculated. And provide it again to the plant. The other thing is plants will drink water to cool themselves.
It doesn't necessarily contribute to that plant's growth, which means it doesn't contribute to the final weight of the product. If you've got the right environment, that plant will always kind of drink its water with the nutrition it needs to continue to grow, and so that water contributes to yield and so you get even more efficiency that way.
What problem is Plenty tackling?
Ryan Schools from Plenty: Our mission statement is to improve the lives of plants, people, and the planet. In our current agricultural system, we've optimized around calories but not around nutrients. And the system that we've designed is hyper-efficient at producing a ton of calories to keep everyone more or less fed, but it doesn't take into account a lot of the externalities that come from the crazy supply chain of shipping food all around the world.
Heavily subsidizing crops that use a ton of fertilizer and a lot of land and water. The company is changing that system to bring nutritious food close to the point of consumption and to produce it in a more efficient way by using less land, less water, and with renewable electricity.
The final piece is the people. Not only by having the food close to them - it's also more nutritious. You can grow crops that can be available year round right now if someone lives in Chicago or on the East Coast in the winter.
All of their leafy greens, berries, etc. Largely, it's coming from a warmer climate. And it's getting shipped and flown through these refrigerated supply chains that stretch around half the world. There's a big carbon footprint associated with all of that.
What problem is Freight Farms tackling?
Jake Felser from Freight Farms: I suspect Freight Farms’ answer to this is different than Plenty. The main difference with Freight Farms is that we have a giant network of small farms rather than a small network of really big farms.
There's a whole bunch of reasons why you might want that or might not want that, and we can talk about those. Freight Farms’ mission is making quality food accessible to people. That's where we got started and where we still see ourselves. A lot of that is less about competing in grocery.
It's more about what are all the places that don't have good food that’re a little bit smaller. And we can compete in all of those spaces all at once because our product scales in a different way than some of the big players. One way of thinking about it is like niche at scale.
Obviously there's a lot of economic headwinds in CEA (controlled environment agriculture) right now, so it's definitely an interesting time to be talking about CEA. Energy costs are high. CEA is generally pretty energy intensive. But this has always been true to some extent - which is that the cost of produce from CEA is relatively high compared to other methods of growing.
But the quality is better as well. And this is true across all the growers. I'm sure it's true for Plenty as well. But for grocery stores and mass market adoption, they're not necessarily looking for that. So how do we position our customers’ products in a way that makes the most sense?
It comes down to the economics that work best for CEA. And this goes to CEA's place in climate. The economics work best when the farm is in a place that is limited in its natural resources, whether that's light or land or water. Those things all make it really hard to grow in traditional ways, and so CEA makes more sense there, but a lot of those places are smaller islands or more isolated markets, and it can be hard to compete at a grocery scale there with a giant farm.
Who are your archetypical customers and locations?
Jake Felser from Freight Farms: So I think food deserts is partially what I mean. But if you think about limitations in land or light or water, for instance, Caribbean Islands: lots of light, but there's actually not much [fresh]water. And there's not much land. So that's a place where CEA makes more sense because the alternative to CEA is not to have giant fields of lettuce like you might in the US. The alternative to CEA is to import and then that's gotta go on a boat, which is expensive and is a big hit to quality. There might be customs or tariff-type considerations as well.
Far north in places like Sweden, you wouldn't consider it a food desert per se, but they have very short growing seasons. So similar constraints apply. We have a lot of farms in Alaska or in Northern Canada where the temperatures are really extreme and there's not a lot of light.
In the Middle East there's not much [fresh]water so CEA works well there. Energy prices there are also lower. Literal deserts more so than food deserts to some extent.
Produce as the product
How do you decide what to actually grow? Is it demand-driven from consumers or is it by what infrastructure and technology can best grow?
Mike Ross from Beanstalk: It's entirely driven by product. So because we're in soil, we can grow anything. In fact, we've done roots, fruits, and even flowers. We currently grow salads, herbs, and micro-greens because we're focusing on these heirloom pre-made salads. And so we make some educated guesses as to what the consumer is going to really want from a product standpoint.
As engineers, we do a ton of testing. But we know the market wants ready-to-eat salads, and so that’s definitely been a resounding success. Now it's about continuing to develop recipes and continuing to improve on all aspects of the experience. But we’re not limited by what we can grow as we can grow anything we would need to.
Why hydroponics?
Jake Felser from Freight Farms: So there are a decent number of studies that show better nutritional content for CEA crops. A lot of that is control and consistency in the nutrient delivery. There's also some more novel crops that are best grown in CEA like the Norfolk Produce Purple Tomatoes.
I don't know if you've seen those, but they're these tomatoes that are super high in a bunch of antioxidants. At the end of the day, it comes down to what can be grown profitably. Really, you can grow just about anything in a CEA setup, Freight Farms or not.
The question is, does the spacing work? Does the amount of labor it takes to put it together work? And if you're not efficient with your use of space, then your ratio of sort of CapEx to what you're selling is gonna be off, and it won't make sense. So you could grow corn in a Freight Farm, but you wouldn't make money.
That's where you get to the leafy greens. Leafy greens and herbs grow really well hydroponically. The cost to sell them is pretty appropriate - you can make money doing it. It ends up being a good business for our customer.
One thing you said earlier which I wanted to touch on, you alluded to, is that it's a human centered product. I think that is absolutely the way to think about a lot of this. The way that Freight Farms works is we sell farms. We don't actually sell lettuce, we sell farms. And most of our customers - what we call small business farmers wanna run a business and be profitable, but that's just part of it. They're getting into it because they care about their community and they want to make food for their neighbors and friends.
Hydroponics vs. Soil-based
What makes Beanstalk different than Bowery or Plenty, where they’ve gone down hydroponics route?
Mike Ross at Beanstalk: We are the first ones to come at it from an infrastructure level and focus on simplifying the infrastructure needed to grow crops indoors. We are a vertical farm. In fact, we're the densest vertical farm in the world. And we did that by inventing a new form of farming.
So we're not hydroponic. We do use soil. But we're able to grow faster than hydroponic farms. We have more nutrition than hydroponic farms. And we're able to go denser because we've removed irrigation. So what happens is we've got these containers that hold everything that the plant needs, all the soil, the water, the seeds of course, and then we can put that into racking that is tended to by a robot.
Basically this farm looks like a giant vending machine. It's incredibly dense and it's got this machine that goes across the face of it, a crane or an automated forklift that pulls things out, pushes things in. So that makes it incredibly dense. We’ve removed about 80% of the components that would typically be in a vertical farm which makes this incredibly cheap. So we're an order of magnitude cheaper to build than any other farm in the world. It is much cheaper to operate because we can perpetually reuse all of our inputs since soil can be reused indefinitely. We currently buy seeds from a variety of vendors all over the globe, but we can produce our own seeds when we get to proper scale.
That eliminates a ton of the costs. And then of course, we develop our own automation. Which are one to two orders of magnitude cheaper than comparable vendors in the indoor farming space. So it's a collection of things, but it starts with our novel farming method that uses this soil container that enabled us to do it.
It created a lot of other problems by going that dense. We had to invent our own airflow system and our own air conditioning system to keep everything uniform. Those were a few big technical hurdles that we had to overcome. But really it is about production density.
Climate Mitigation vs. Climate Adaptation
How would you put Freight Farms in regards to the spectrum of climate mitigation vs. climate adaptation?
Jake Felser from Freight Farms: Yeah, so I think all of CEA I would put pretty squarely in the adaptation category. And I think this is a mischaracterization that's pretty common in the broader industry that people get really focused on food miles.
But if you look at the LCA of a traditionally grown head of lettuce, food miles don't dominate. It's still dominated by water and sort of other resource uses, not the food miles at the end. So people get hooked on the food miles, but CEA is so energy intensive.
And so few CEA facilities are using renewable energy that it's hard to make a mitigation argument today. You can definitely make an adaptation argument, which is that in 10 years or 20 years we still want to eat vegetables. We better work on the technology that we're gonna need to do that.
And I do think part of tackling the climate crisis is also recognizing that we all have a role to play. I'm not an expert in nuclear fusion but there are people who are. So if they solve that problem, then all of a sudden CEA is a mitigation strategy. Because in a world with cheap energy or cheap from a carbon standpoint it's a great option because it's much more efficient in water and chemicals in general land impacts - land that could be used for sequestration or other things.
So that's kind of how I see it. I think today it's largely adaptation, solving the energy problem in parallel. It's both. Yeah.
The Role of Capital
If you could wave a magic wand, what's something that would make your life easier and help accelerate the business?
Ryan Schools from Plenty: If we could get a bunch more funding fronted from partners, we could go out and build a bunch more farms. The more that we can scale and have higher demand for all of the equipment that goes into these facilities, the cheaper all of those pieces of equipment are gonna become and the easier to manufacture them at scale.
Right now we're on a pace of building and deploying a farm every two to three years. It's very batch based. All of our partners that fabricate our equipment get turned on for a year and then they sit and wait until we're ready to deploy the next one. Just like any early stage industry that is coming up the scale curve, there's a breaking point where you can switch from batch-based to continuous manufacturing of all these inputs, and you're gonna see pretty dramatic efficiencies come out of that.
It's difficult as a 400-person startup for us to go out and continue to raise the $200-300 million it takes for each of these facilities to go and build them ourselves, right? It's kind of like the airline industry. Boeing goes out and they get a huge backlog for hundreds of jets that they have on their order books, because jets are really expensive and complicated to build.
If we can do the same thing and build up that backlog where we have guaranteed work for years and years, then we can go out to our suppliers and show them that and basically just turn the lines on and start building all the stuff that we need to go and make those farms.
Interconnectedness
I remember I asked you where you would put the farm if you could put it anywhere and you said wherever energy is the cheapest. Something I've been learning about climate is how interconnected things are. Can you say more on the interconnectedness in this space?
Jake Felser from Freight Farms: Everything is interconnected. It's honestly hard to overstate that. So energy pricing is interconnected. Energy pricing is a big lever on a lot of things. For a lot of the bets around home electrification and broad electrification, there is an efficiency gain there. Your actual energy input for an electric car is lower because it's more electric drive. Trains are more efficient than ICE drive trains, for instance.
There's actually a Sankey diagram for savings and doing that. But if you're creating electricity from coal, it's not necessarily better from a carbon standpoint. So a lot of things do go back to energy. If you had limitless clean energy you would solve a lot of issues because you'd basically say, well, anything electrified is now good.
But energy is tied to the grid. That's a bet that says we have actually electrified enough stuff to make that work. If you electrify more and more things, you have other grid impacts, those things are tied together. There's software that controls everything. There's all these endpoints to the system and there's all these systems that could be more efficient.
For instance, if energy was free, then do you care about HVAC efficiency or do you just care about whether your refrigerant has a high GWP? If energy is free, maybe you don't care about efficiency so much anymore in a lot of places. So everything's interconnected and I think the challenge of that is it's hard.
It's hard to focus on any one problem. To have an impact on that one problem, you have to focus on one problem and you have to understand the threads it pulls on. But let's say you spend 20 years focused on efficiency and actually the right thing to do is to focus on electrification. That might be a wasted effort to some degree.
The other danger is that you get paralyzed by all the interconnectedness and don't do anything. And there's a third danger that you do something, but it's interconnected in a way you don't see and causes a negative impact. But I think the opportunity and things being interconnected and the bet we all have to make is basically I solve my part, you solve yours, and we have to talk enough that lets us all stay aligned and make sure that there's no interconnections that are gonna bite us.
HVAC
I remember you talked about heat pumps because you have to think about HVAC and temperature control and humidity for Freight Farms. Why did you even have to deal with that?
Jake Felser from Freight Farms: Yeah, so I mean, HVAC systems for indoor farming are very unique and different from sort of traditional residential HVAC because there's a much, much higher humidity load. So Freight Farms has custom developed an HVAC system for our farms that is different to go back to the interconnectedness element here. Where it gets really complicated - you touched on labor shortage earlier - that's definitely a part of it, right? If you want to build a custom HVAC system and work with a good vendor to do it, there's not that many places to do it.
Then if you want to service that system in the field at scale like we do, we have hundreds and hundreds of HVACs out there. You need a network of support technicians. But what we find is that the quality of the service that you get is wildly dependent on what technician you call. So and so you end up relying on the manufacturer for that.
Does that scale? I was talking to a friend of mine at a party over the weekend. Who works at an HVAC company in operations in New England. I was just talking about pricing and I tried to get a heat pump in my house in the fall, and the pricing was just prohibitive. And so I was asking him about it and he basically said that whoever gets the most trained people wins, because that is the primary constraint in the HVAC market right now from his perspective. So how does that all kind of tie back, right? I mean, Innovation in HVAC makes it. Without innovation in HVAC, the CEA doesn't work, right?
You basically need those innovations to happen. So you need trained HVAC techs out there. You need people in the trades who are gonna build stuff and install these things. All of that stuff has to happen, even if you're training people today who aren't necessarily installing electrified appliances off the bat.
And then there's a regulatory tie-in too. So if you look at some of the regulations coming out around heat pumps, there's regulations that are gonna come out in California, for instance, that require heat pumps everywhere. No more one-way air conditioners. Now the interesting thing about that is that makes total sense for residential, especially in California where it's temperate and you're gonna have to heat and cool so you get both with one shot. Makes sense. And the cost is coming down. But for a specialized system in a farm, it doesn't make sense because we're always generating a lot more heat and we're always gonna have to dehumidify. So we're already looking at some of these regulations and trying to figure out how to engineer around them. But we're a small company and we can't go lobby, for instance.
Matt: This is an interesting example of a pro-climate bill negatively impacting a company that is also trying to do good for the climate.
Jake Felser from Freight Farms: Yep, totally. Yeah, it's an example of those interconnections where if you're not thinking about it fully or all the way through you can run into trouble. But in the grand scheme of tradeoffs, complexity of legislation isn't necessarily a great thing. So do you wanna make it super complicated with a lot of carve outs? Do you wanna not have the legislation? Probably not, because I think it's really positive for residential. So you have to weigh the impact of it.
Standardization
How variable or unpredictable is the growing process? I imagine there’s some formula of water, sunlight, temperature, etc. If you run that a million times on the current process, is it gonna look very different or is it fairly reliable?
Ryan Schools from Plenty: In my opinion, that's one of the biggest challenges - how do you standardize everything when you're dealing with a living product? Everything is not going to be the same, even with the same inputs. You can hope to contain the distribution so that things fall within a certain tolerance band. At the end of the day, it's not manufacturing, not non-biological hardware, right? Plants are living and they're complicated biological systems.
How can you select seed cultivars and place your set points for all of these environmental factors like the light load, nutrients, water? - all these things that the plants need to grow. And then dump all that into the black box that is the plant system and get a consistent output that doesn't mess up your automation. Because you want the plants to be the same height after the same amount of growing time so that you can have an automated robot come and cut them all off for harvest and not have leaves that are different sizes or shear incorrectly because nobody wants to see that in their box of leafy greens that they're buying at the grocery store.
It's a huge problem and one that's not fully solved. Leafy greens are one of the easier ones to solve that problem for and that's why they're kind of the first to market. And then, I think more and more products will be coming as that equation gets fleshed out a little bit more for other types of produce.
What do you believe about this industry that's different than everyone else?
Mike Ross from Beanstalk: Our focus is on nutrition and human health. We're focused on developing products that make it easier for people to eat healthier and tastier food. I think a lot of indoor farm companies have focused entirely on innovation for innovation's sake. I think that's been the reason for a lot of the companies that unfortunately didn't make it. We put it simply: we have three success criteria for the business: to improve the taste, convenience, and sustainability of fresh produce. If we improve the taste, you'll crave it. If we improve the convenience, you'll default to it. And if we are a sustainable process, it'll be around forever. And that's it. That's all we need to do for this to become a true success.
Conclusion
I hope this was useful! My key takeaways were:
Everything in climate is interconnected, from energy to skilled labor to HVAC to food production.
Indoor farming solutions tackle a wide range of problems (nutrition, food deserts, climate adaptation, traditional agriculture) with a variety of solutions (hydroponics, soil-based, the farm itself)
Indoor farming as a climate adaptation solution is not how I initially thought, but I find it compelling now.
I didn’t expect nutrition and the human-centric view to be as emphasized as it was. Hopefully we can all get more nutritious, nourishing food, whether it’s grown on the same ole’ flat land or in a big manmade box.