Farmers typically consult the calendar and the weather forecast to figure out when to plant their crops, but figuring out how to grow a tech startup focused on the farm can be a far more complex task.
The challenge can call to mind the old joke about the farmer who won the lottery. When asked how the winnings would be used, the farmer answered, “Well, I guess I’ll just keep farming until the money runs out.”
When it comes to ag tech ventures, the money isn’t running out: Last year, a Crunchbase survey found that venture capitalists were investing roughly $4 billion a year in farm-centric startups — and the flow is continuing despite the COVID-19 pandemic. So far this year, investors have put about $700 million into more than 90 ag tech ventures, according to Crunchbase’s tally.
Some of the stars of the show are Pacific Northwest entrepreneurs who found success in the software industry and are now bringing their startup savvy to the food and agriculture industry. We checked in with four founders to get a sense of how they’re cross-breeding technology with agriculture:
- Brent Frei was a co-founder of Smartsheet and the former CEO of Onyx Software. Now he’s the founder and CEO of TerraClear, a venture that’s developing rock-picking robots for farm fields. The prototypes are being tested at the Frei family farm in Idaho.
- Paul Mikesell’s best-known venture, Isilon Systems, was acquired for $2.25 billion in 2010. Just last month, his latest venture — Carbon Robotics, which is building a self-driving, weed-zapping robot — came out of stealth mode after years of hush-hush development work.
- Keith McCall, who co-founded Azaleos and Enroute Systems, has since gone on to create a robotic crop monitoring startup called Pollen Systems. This year, he acquired and relaunched Everyvine, an online marketplace for wine grapes and other vineyard products.
- George Chrysanthakopoulos, a former Microsoft engineer, founded Directed Machines — which started out building a laser-equipped weed zapper and is now marketing a solar-powered robot that can take on land care tasks such as mowing, trimming and grading.
Each founder responded independently to the same set of emailed questions about the state of ag tech. We’ve lightly edited and condensed their answers:
Why are software entrepreneurs going into ag tech?
TerraClear CEO Brent Frei: “It’s a massive market both in real terms and in its opportunity to directly improve one of the most fundamental human processes. There are fewer and fewer farmers around the world trying to feed more and more people. And they’re doing it while trying to keep the cost of food low for everyone and at the same time remain excellent stewards of the environment. Everything we do to increase productivity while improving environmental impact is super helpful worldwide. On top of being a very good business opportunity, applying cutting-edge technology to ag is incredibly fulfilling, particularly when you’ve personally lived it without that technology.”
Carbon Robotics CEO Paul Mikesell: “Software entrepreneurs love getting computers to solve problems. Getting computers to solve physical-world problems is even more interesting. Getting computers to control 1,000-pound robots to work on something as fundamental as food supply — amazing!”
Pollen Systems CEO Keith McCall: “I think many entrepreneurs are seeking a new frontier and the agricultural industry is definitely behind in terms of applied data science and software solutions to address a variety of challenges from improving crop health and growth to reducing pesticide utilization and conserving water. The opportunity to both help people with the basic need for food as well as the opportunity to help the world at the same time is too much to pass up.”
Directed Machines founder George Chrysanthakopoulos: “For me, it was to remove pollution of all forms — noise, air, light — and to help small diversified farmers and rural property owners (I am my own customer).”
Why did you choose the ag tech venture that you’ve chosen? Is founding an ag tech venture fundamentally different from traditional tech ventures, or pretty much the same?
Brent Frei: “It’s extremely rewarding to bring cutting-edge tech into one of the world’s oldest and most meaningful professions. I grew up on a farm, and absolutely love working with farmers as customers. Farmers are amazing people. TerraClear (ag tech) is very similar to Smartsheet (traditional tech) in that they both solve big problems in giant markets where a winning solution has been forever elusive. In starting companies, I have always begun with a hard problem. This process of staying laser-focused on something that helps the customer is incredibly rewarding. And that is even more true when the customer is a farmer. It’s an easy thing to get good people behind.”
Paul Mikesell: “We spent a lot of time with farmers, listened to them, and worked hard to really understand their organization and input costs. These farmers are really savvy business people — they know what’s costing them money and causing problems. In many ways they’re really an ideal customer — show them an ROI and they can justify spending immediately. It’s not like you have to go through some corporate CIO organization to get purchasing approval. The guy who you just spent the day with in his fields, who loves your solution — he also signs the checks.
“Founding an ag tech venture is different in that you really do have to spend a lot of time out at the farm understanding the environment, and testing testing testing. One farmer told us early on ‘whatever you think is tough — double it!’ Well, he was right.”
Keith McCall: “After selling a couple of other startups (Azaleos and Enroute) to larger organizations, I took a year off at the request of my wife from being an entrepreneur and traveled the world. When we spent time in Singapore I was struck by how that city is using technology to address climate issues and to try to mitigate their carbon footprint by planting Supertrees in a large grove. Applying data analytics and aerial imaging to help grow better crops — which we can now track down to the individual plant level — became the founding principle of Pollen Systems.
“Agriculture is very different because of the growing season and the sales cycle. We built companies in both the USA and in Chile to capture the growing season throughout the year. And the sales cycle predominantly takes place in the off-season in each geography. Those two factors are fundamentally different from any other startup I’ve built.”
George Chrysanthakopoulos: “I started with laser weeding / directed energy, which is now becoming popular. We soon decided 10 different robots doing 10 different tasks is going to drive our customers mad, so we did something much more challenging. One robot to (mostly) rule them all: weed, haul, tow, ground engagement, all with three different types of autonomy for both structured (rows) environments and pretty much any outdoor space. It seems to be paying off, since it’s the same core tech hardened to deal with massive variance. Making it affordable was key, another massive differentiator. Still early though!”
Are there lessons from the software world that were particularly useful for a hardware-centric startup? Any lessons you had to un-learn, or new lessons you had to pick up?
Brent Frei: “The product has to be very easy to use, and has to work as expected every time. In a software application, an alternative product is one click away. In ag, the effort to demo an alternative product is very expensive, so farmers are far less forgiving when something doesn’t work. They never give it a second chance.
“That said, when farmers trust a company and know they are trying to solve a real problem, they are very interested in helping. We have always maintained a rapid iterative design process and it is chock full of farmer feedback at every level. I’ve been pleasantly surprised at just how much time and insight farmers have been willing to provide TerraClear because they believe in our goals.”
Paul Mikesell: “I’ll tell you one tenet that’s important for robotics in general — whether you’re talking about ag robots, or drones, or delivery robots: Get your machines out into the field! Too many of these robotics companies fail because the engineers spend all day behind their desks designing and coding. By the time they deliver what they think is the ‘perfect’ machine they realize that things are much different in the real world than they imagined, and they wasted a bunch of time on the wrong things. This kills more companies than people realize.”
George Chrysanthakopoulos: “I always did stuff with small teams (myself writing most of the code, still the case), so kept doing that here. So do more with less. Big budgets early are bad. I said in my 2018 article to be patient, ignore the hype, build something useful. I still think that is a solid approach.”
In the post-pandemic startup environment, is doing an ag-tech venture harder or easier than in the “before times”?
Brent Frei: “TerraClear has had a more normal working environment than most companies, because a portion of our team spends much of their time outdoors. Farmers work outdoors all day long, and for the most part haven’t changed their farming practices during the rest of the world’s lockdown. That said, much of our team has worked from home. We have evolved like every other company over the past year, and in many ways have become more efficient. We actually feel empowered at this point, more flexible and nimble to shape our workplace environment than we were pre-pandemic.”
Keith McCall: “Ag tech companies require interaction with the grower or owner of the field. Although Zoom has been a great tool at bridging that gap, we are eager to get back out there and meet the farmers face-to-face. More than many industries, the agriculture vertical takes time to break into and patience in developing relationships over time, while at the same time growers face a new distraction every week during growing season.”
George Chrysanthakopoulos: “It’s super hard before and after, but some types of labor are not coming back, so automation is easier for customers to accept.”
Where does ag tech fit in the wider landscape of innovation and entrepreneurship? What technological frontiers look most promising down on the farm?
Brent Frei: “Farming is far more automated than most people understand, but some of the hard jobs in farming do remain unsolved. Repetitive tasks or those prone to human error are excellent targets for automation. Technologies in search of a problem will always have a tough road to adoption on farms, but technologies that specifically reduce manual labor, reduce input cost, reduce the chemical load and increase yield will be very welcome.”
Paul Mikesell: “The way I look at things is like this: Imagine what you think things should be like in 100 years. Then figure out how you can participate in getting us there.
“When I think about this, it’s clear that robotics and automation should be used to remove a lot of the hard human labor that’s dangerous, repetitive, tedious. Isn’t this exactly what we’re doing with most of our software systems today? We just haven’t deployed these kinds of techniques to the physical world because of the lack of robust computer vision systems, and we need good use cases to get started.
“Warehouse automation, construction, and ag are just a few of the beginning use cases for a whole host of future automation and robotics solutions that are going to bootstrap the next revolution in tech innovation.”
Keith McCall: “I considered doing something in one of two frontiers: something in space or something in agriculture. Both are critical to the future of our planet and our species.”
How will tech innovations change our image of farming and farmers?
Brent Frei: “Most people are pretty clueless on how farming works and how their food is produced, but that seems to be changing quickly. The tech in ag now is fascinating, and that will inevitably bring curious people closer to the industry. And it’s making food continuously more affordable and continuously greener, so how can you not be interested in that?”
Paul Mikesell: “There’s already a revolution in automation happening on the farms. A lot of people just don’t see it yet. In an couple of years the farmers are going to have some of the most advanced autonomous technologies running their operations — and the best part is that they’re doing it in a way that’s going to save them money and has a real ROI associated with it, which means it will have this flywheel effect on the innovation cycle. It’s not just a fad.” reduce the chemical load and increase yield will be very welcome.”
George Chrysanthakopoulos: “We do not respect or appreciate how hard it is to grow food, in a small farm, in a decentralized fashion, sustainably. I do not target industrial ag like a lot of my competitors because I think it’s way too rough on the soil (intense!). So I am hoping we make small-scale, local food growing, land care economically viable again, and bring more people into it.”
These four companies are just the tip of the iceberg when it comes to the Pacific Northwest’s ag tech startups. Here are 10 more ventures that are bearing fruit:
Got additional ideas for Northwest ag tech coverage, including ventures we’ve missed? Email your suggestions to [email protected] with “Ag Tech” in the subject line.