Providing growers more precise and accurate knowledge about their orchards

In this week's episode, Thomas Slattery spoke with Jim McDougall, Commerical Director at Outfield Technologies about their exciting technology helping top fruit growers around the world.

Jim is hugely passionate about the future of fruit production. In his previous career assessing environmental risk, it became ever more apparent that we need to become more efficient as a species at feeding the population whilst protecting the planet. Whilst there was a lot of talk and good sentiment, action was needed to really drive change.

And Outfield was born. 

At a time of high pressure in the farming industry, Outfield is focussed on helping fruit growers be more productive, more efficient, and more sustainable. To achieve this, Jim and Oli founded Outfield with the ambition of providing growers more precise and accurate knowledge about their orchards. After years of working with growers, researchers, computer vision experts and partners across the industry, Outfield now has a system being used by apple growers across 5 continents to manage orchards with precision, forecast yields with accuracy, and inform the whole supply chain.

Transcript

0:03

Good afternoon. And today we're speaking with Jim MacDougall at Outfield Technologies. Hi, Jim. Hi Tom. Great to be here. Thanks for having us on. Not at all. Absolutely a pleasure. So, first off, for anyone listening who doesn't know about Outfield, do you want to just give us a little bit of a kind of background to the organization and what you're doing? Absolutely. So, Outfield, we really believe that the future of agriculture lies in agri data. I can sort of explain more about where that came from, but what that means for us is that growers really need to understand what's happening in their production so they can make precise interventions, so they can make the most food possible using the least inputs. So we've spent several years working towards product that we now got at market, and that's focused on fruit sector. So there's a 200 billion dollar sector where we're dealing with big three dimensional canopy structures where growers will send people to go and look at maybe ten trees in 10,000 to try and assess how many fruit they're making. To try and make decisions about where to deploy agrichemicals. To try to maximize their yields and to try and prime the whole supply chain by forecasting what they'll make. This obviously doesn't give them particularly good results because although they do amazing things with that very limited data point, they really need more data to be making better decisions. So what we do is we ask them to buy an off the shelf drone. It's their drone. They can get one on Amazon Prime anywhere in the world tomorrow morning, and then we send that drone a flight plan so they can walk into the orchard and press two buttons. They can survey their crops in 20 minutes. They send us pictures of the trees and then we analyze those using amazing machine learning systems built by people out for a much cleverer than I am. Which then tells them how many fruit they've got per tree or how many blossoms they've got per tree. Tells them about what their yields are going to be. And tells them about what interventions they should make to maximize their outputs and to minimize the sprays and fertilizers agrochemicals and irrigation that they need to get what they need out of the auction. This also primes the whole supply chain because at the moment, yield estimates tend to be off by about 20% even on the day of harvest, allowing them to sell with confidence and sell at premium prices. Get the fruit where it needs to be in the world. We can also I'll leave it there for now. No, absolutely fascinating. So obviously you're focusing primarily on the fruit sector at the moment. I'm sure we can come on later where you see yourselves going in the future. And first. I find it really interesting that not only you helping the growers. But of course. As we know. Particularly in the fruit sector. The specific size of fruit is a big thing for the retailers and giving growers the ability to help forecast that not only allows them. I'm guessing. To work most effectively with their buyers. But also plan what they can do with fruit that might not necessarily fit into the quite strict size and quality requirements. Absolutely. And there's a real catch 22 in the industry that around about the time that they know what they'll have by the end of the season is around about the time they can't make any interventions anymore. You can do things to change the size, the grade and the quality of the fruit, but by the time you know, it's usually too late. So that's a tool that we're really giving growers to be able to get in early to make interventions. And obviously this exercise has so many pressures at the moment that the more efficiently they can be doing things, the closer they can get to being profitable. Everyone's better. Yeah, I mean, just quickly on that. So obviously, I think we all know that the labor crisis, however we want to term it, is something which is an ongoing problem, is going to be, again, a problem this summer is that going forward is being able to use your technology potentially in something like helping to more optimize where pickers are going. Is that an area that you're looked at? Yeah. And labor is definitely something that we can help with. It's not our primary focus, but knowing, as you say, where your fruit is, how much fruit you've got, deploying that labor most efficiently and most effectively, having all pre planned, having your bins to pick in the right places, all that's really valuable. But where we're putting a much bigger dent is around things like precision spraying, which reduces the need to have labor interventions throughout the season. So if we can get blossom thinning exactly right, using chemical applications, then you don't need to send people around to do fruitless thinning, which takes days and costs thousands and thousands of pounds per hectare. Yeah. Do you want to explain a little bit about the kind of theory and the blossom thinning and how that has an effect on kind of fruit count and fruit yield? Absolutely. So I'm not an agronomist, so any agronomists listening can respond in the comments. But in a very brief. Apple trees in particular are biennial, which means that they turn on and off seasonally. One year they'll produce a lot of blossoms, which means they produce a lot of fruit, and the following year they'll produce very few. But it's very unpredictable. And different trees in the orchard will respond differently. In fact, if you leave an orchard alone, they tend to go into a zebra strike pattern where they'll have trees on and off next to each other. Now, what this means is for a tree that's got too many blossoms, they'll wind up with too many fruit on the tree which come out too small because there's not enough energy to grow the fruit. And the reverse applies as well. Exactly as you said, we're aiming a very narrow band for what size fruit we want and that's not driven by supermarkets being dastardly, it really is just what consumers will buy. Here in the UK we have a band that's smaller than the rest of the world where we like slightly smaller fruit because that's just what we're used to. Whereas if you go to the continent you'll see they prefer much bigger apples and that's what they have to aim at. So if we can get the right number of blossoms on every tree early in the season, gives you a nice homogeneous orchard, the same all the way across. Not only can you treat that all as one complete thing which allows you to be more efficient with how you're managing your orchard, but it also allows you to get more fruit on grade on the correct size, quality and color at the end of the season. Yeah, it's amazing. So I just want to pick up on something you mentioned at the beginning which I think is really important and that was the fact that you developed your technology to be able to be used with off the shelf drones. I think one of the things that we've seen at Agri-EPI a lot is making sure, the importance of making sure that farmers at all levels are able to who are interested in using these kind of technologies are able to access them. And of course, when hardware comes into play, having customized and very expensive hardware can be really limiting for that, particularly with kind of cutting edge technologies like this. So obviously you've made this very wise decision to make yours available for off the shelf drone products. That kind of leads into lovely to hear a little bit about how outfield got started and what the kind of story is in terms of development where you got to this place. Now absolutely all credit for that idea goes to the other founder of Outfield, Oli Hillbourn. Who's an aeronautical engineer by background and now obviously been running outfield for six years. Does a lot of work on the development operational side and he was very aware very early on that in order for these kind of solutions to have the impact that we want them to. It's got to be impact through business. Not impact or business. So it's got to make commercial sense. It's got to be scalable and also it's got to be scalable not just in the first world where we've got the ability to deploy big robotic systems in field or pay for train pilots to come out and use specific systems. But it's also got to be deployable in the Third World or in developing economies where we still have fruit production. South America is a huge fruit growing region. They might not be willing to deploy that kind of technology. So he planned right from the very get go that all of us be working using off the shelf hardware. Hardware is hard and the drone technology has really burst forth over the last decade or so. If you go back ten years, no one knew what a drone was. Now your son's probably got one, right? And so because of this, I mean, growers we speak to all around the world already have a drone that they can just deploy right away, sitting on their shelf. And so we just find it good use. On the other side of things where the idea kind of formulated was I was working in private equity, looking at people buying and selling things like chemical facilities. And so we're trying to assess the impacts, the environmental impacts of, say, farmland around them. I'd often speak to farmers all around the world and say, what did you spray last year? And they say, hey Bob, what did we spray last year? I'd realized that they're talking about a$10 million facility that deployed half a million dollars of chemicals last year. And Bob might know where they put those chemicals. The cutting edge technology in some parts of the world really is the pen and paper. And so with all these understanding of what was possible with this technology and what you could do with surveying techniques and my understanding of what was really happening on the ground, we sort of put these ideas together and started to explore ways of getting more data out to growers to make things more efficient. We had a quick look at our history and looking at previous agricultural revolutions and what happened in the with the Green Revolution, making more food, feeding more people, but in that case, using a lot more agrochemicals and using plant breeding. We're now facing the same challenge again. But it's even harder this time because we've run out of horizons and we're running out of the ability to deploy chemicals. We've now got to do even more using less. And that's why we really think that data a lot like a lot of the other people that you've spoken to in the podcast, actually, we really think that agri data is the future of this whole sector. Yeah, I think I would agree with you a lot on that. But what I really love about what you and some of the other organizations, for example, drone AG who are working very similar but in Arable and Sensibly, also built their technology to work with off the shelf drones, is again making this stuff easily accessible for farmers to use without having to become technology experts or spend a lot of money and so on that you guys have. Thankfully, you're commercial now. It'd be really wonderful to hear about maybe some examples of growers that you've been working with and how the technology has been helping them. For sure. Yeah. So we are commercially, as you say, and we've got clients sort of on a trials basis, deploying this around the world this season with Views are really going to a big international launch next year. Everything we're doing now is scalable, everything works for a pipeline. We're processing orchard data without having to touch it, which is fantastic. It's a huge change for us. For the last couple of years in the UK, we're working with the UK's biggest grower, which is AC Goto. We've been working with some very forward thinking, fantastic growers like AC for several years now and working with the research bodies as well, which is great. And also working with the producer organizations like Avalon. They know, they do the sales and marketing for a lot of growers, so they really care about this data as well and understanding what's coming down the pipeline. We've also now got growers in Michigan, New York, in Italy and South Africa, New Zealand, Australia and Chile. So although sort of early tests and trials in different markets, it seems like there's a real pull to get this technology out there. And just last month in South Africa, we started embedding the data that we are producing, the maps that we make into precision spraying robots. This is another sort of step change for what growers can do with this data that allows them to put the right amount of agrichemicals, the right amount of fertilizer, in this case down in the right places to deal with the variability that's in the orchard, to produce the right amount of fruit and get it on grade without having to spray where they don't need to. For these growers in South Africa in particular, the cost of fertilizer has gone up four and a half times in the last year, which means that it can take their orchards from being in the black to being in the red. They either have to choose to not spray and then they lose quality, or to spray and then they lose cost. So we can help them in this case put down half as much fertilizer in half the time, really improving the profit margin. That's really interesting and we might be going off a little bit, but you talk about a precision spraying using integrating your data into robots, which I think is a really neat solution. Obviously in the UK we aren't able to use drones themselves for spray applications. Now this is something which we as an organization Agri-EPI, and yourselves are speaking with HSE and others to try and work out how we can kind of shift that policy. But you're in an enviable position now, working internationally and in both hemispheres, which obviously doubles the amount of seasons that you get to test your technologies. Great. Is that an area that drone applied spraying, is that given that you're working in markets where that's potentially more lenient to that, is that something you're excited about? Is that an area that you see things going? We really strongly believe and always have, that there's not going to be one company that wins this space. It's going to have to be a collaboration with other people. And there are people out there who are doing amazing things with drone spraying and we're really happy to sort of defer to them as the experts on that. Feeling on it, at least at the moment, is that drones are very good at carrying light things like photons, so they're very good at taking pictures. But once you start carrying huge amounts of agrochemicals around, that's less there forte and the time pressure is not the same there either. So we think for certainly the big sprays for doing things like for putting down fertilizer or PKN, things like that, that's probably better served by having a ground based system. The risk is lower, the complications are lower, and it can just trundle through the orchard then I can do that. However, for some things like targeted interventions for pests and disease, that's a really hard problem and we don't think anyone in the world is even close to doing good pest and disease detection just yet. Once we get there, I really think that is a good application of targeted drone sprayers to go and take out eight trees worth of cuddling moth or whatever it is. Do you think that the detection is because there just aren't large enough data sets or it's too hard to kind of capture the right imagery even with drones to detect that over kind of large orchard sizes? In our background as a company, we started out doing sort of consultancy, working with plant scientists to do all sorts of machine learning solutions. So we had a bit of a dabble with sort of early disease detection stuff. There's several big barriers to how that technology sort of needs to get to in order to be effective and efficient. The first biggest one is how you actually detect a disease. And if you look at how it's done at the moment, agronomists walking around the ocean really are kind of waiting to see the trees being damaged before they make an intervention. You can catch some diseases early and there are signs of them, but it's really down to luck. So it's about having that incredibly high resolution data, which is assessed at almost a leaf by leaf level for things that are incredibly subtle to notice. The other thing that we've definitely noticed is there are indicators of things like plant stress and so you can use things like NDVI, which we've been talking about for almost two decades now, using infrared lights or thermal cameras, things like that. But then your big problem is diagnosis. So although you might know that half a dozen trees are stressed. I mean, the incredibly multivariate inputs to an orchard can cause that through heating soil, wrong applications of chemicals. You have just trees being out of whack for a year, pests, diseases, all sorts of things. So getting really down to what's causing tree stress and then getting into what pest and disease control you need to deploy. It's a very hard challenge. And I have a lot of respect for people who are out there trying to do this. And actually, we're talking with Agri-EPI at the moment about a grant application to do exactly that. But there's a lot of ground cover, I think, before that's. A commercially ready solution. Yeah, no, definitely. And I know we're excited about the application. So just to go back again to talk about data, I think we're in agreement that data is so important in this sector and again, not just at a farm level, but all the way through the food value chain sometimes. This is a bit of a hotly debated topic and an area that we've seen is farmers can't make a lot of use of raw data and there are a lot of solutions out there that are providing farmers with a lot of raw data. How at Outfield are you working with farmers? You're obviously making it very easy to deploy and very easy to collect data. What are you doing to help give farmers kind of insights and then potentially consultation on that so they can work best, work with agronomists and others and themselves to kind of find solutions? I really agree with this, by the way. I think it's a perfect topic, I think, for anyone who's listening, who's sort of starting or working on an agritech solution. One of the best advice we have been given is to spend more time with the clients. You are not your client, right? You cannot know, even if you might be one of your clients, what all of your clients want to do. So that's kind of really where we started with this. We're spending quite a lot of time with growers, first in the UK and now in other places, really gain to understand where their pinch points were with the type of imagery that we gather, sort of looking into the orchard at a 45 degree angle and taking 200 trees at once. We've got incredible resolution and incredible depth of data in that imagery. And so there were many different things we could have focused on to try and assess from that. Is it going to be foliage mapping, tree height? Is it going to be bigger assessments? And so, by working with the growers, we focus and focus and focus again, down to getting the fruit assessments right first, because that, for every grower, is the data point that they care about. And you'll talk to a grower any time of the year and they'll tell you what they're worrying about. Right now, I'm worried about my blossoms, I'm worried about my fruitlets. I'm worried about my spraying, I'm worried about my harvest. But all the way through they're worried about what my yields are going to be and what my fruit is doing. Because that really is where this all comes down to. So for us, we decide to focus on a base point that the growers are already trying to get and getting it to them better, getting it to them more efficiently, getting them with confidence, getting with consistency. And that goes for the fruit, that then goes to the blossom and the yield assessments as well. And that's really where we focus to start. Yeah, I mean, it's lovely to hear that and it's not surprising given that we work together. But at a core of the philosophy at the Agri-EPI Center is exactly what you've talked about there. Which is making sure that we are working as closely as possible with the people at the front line. At least the farm level. Which is farmers. Producers. Growers. Stockmen and making sure that they are explaining to us what the key challenges are for us both at the Agri-EPI center and as tech developers. Academics. Innovators. Entrepreneurs to most appropriately find innovative solutions to those problems and trying to keep that relationship as close as possible. And I think the most successful organizations that we've seen are the ones that kind of put that at a core as you obviously have done. Absolutely. And I guess I know a lot of your listener base are already in this sector, but I guess anyone who's not or is coming into it. We've definitely found that growers and farmers everywhere really are business focused because they've got to be the ones who aren't go out of business. They operate on a single digit profit margin. On a good year, they make 9%. On the a bad year, they make -2% but as soon as you can get that aligned, as soon as you can get sort of into the same mindset of making sure that you're improving their business, we found growers to be very giving with their time, with their resources, to be very visionary and forward thinking. I think that stereotype of growers being kind of backwards and crotchety old men is not really true anymore. I think certainly the ones that are still around tend to be very business focused, but also very technology focused and very excited about the future. Yeah, definitely. That's certainly something that we've seen as well, both within our satellite farm network and also the wider kind of farm tech circle community that we've been fostering. And also I'm sure you were at Groundswell recently you have to go somewhere like that to see that you've got a lot of the farmer community that are both technically savvy but also looking towards kind of how these things can help sustainability and regeneration. Which would be a nice segue to have a chat about how you see outfield technologies helping growers with kind of pushing forward toward more sustainable systems, definitely. And we really think that this is the direction the whole industry is going. Certainly in the horticultural sector we're seeing now, there's been a lot of mergers and acquisitions of different orchards and we're seeing sort of orchards becoming larger and bigger players in the space. And that's partly for business reasons to make them profitable. But that means that growers often now have a data lead or an innovation lead in house, which is something that's only been true in the last ten years or so before that would be completely unthinkable. And that really ties into where we're taking the business and where we're hoping to drive the sector in general. We've got this kernel of truth that we really think it's at the right at the center of what's happening. As I said earlier, from the imagery we get, there's so much more that we can extract. So really building out a solid understanding of what's happening in your orchard, we can now start pulling in other data sets as well. And so we're bringing in things like weather data on a grant funded project right now. And we've got other ones that we've got in the pipeline that are looking at bringing in what you've been spraying or bringing in data about sort of what you've harvested and what the yields were previous years. Soil data as well. By putting this all together, although I don't like the phrase much, we're really working towards a digital twin of the orchard. The idea here is not just to tell you what's happening right now and not even tell you what's happening in the future, but also to help you to assess what event interventions you could make to optimize different factors. So if we're talking about optimizing your profitability, if you put down certain chemicals, if you put down certain treatments, what will the impact on the end of your year be? And then we can also start looking at including other objectives in that. And as you mentioned just now, the sustainability is really a big driver. And I think you were talking to the guys Agrisound last week about this too. Great podcast, really enjoyed them. You were saying with them exactly the same thing as you were talking about last week with AgriSound. There's a real push from the consumers and from the industry to get better control over environmental damage and environmental impacts. And so that's something that we can really help growers to do, to really help them sell their fruit, to put themselves in a better position in the global context. Yeah, that's very exciting. So what are the next big projects? You mentioned a couple of grant funded projects you're working on. I believe that you've got some potentially some funding rounds coming up. It'd be interesting to hear what kind of the next big steps are for Outfield. Absolutely. So having done this year, having done this season, we've had fantastic results from our growers. We've done several hundred orchards across five continents now, and a lot of that's just running itself automatically, which is amazing. And so, in terms of where we want to take that next, we're doing a big commercial international launch during next year to really get this technology to as many growth as possible, to really help them to manage their crops more effectively, be more profitable, be more efficient, and be more sustainable. A big part of the funding round that we've got coming up, we're looking at raising £4 million by the end of this year to really drive that international growth, to put a presence in various different parts of the world. But the other side of that is that we are also building out more technology on top of this. So, having really nailed blossom mapping, fruit counting, and yield estimation, there's so much more that we can do with the data sets that we've got. So now we're building in fruit sizing and some color assessments. We're looking at bigger assessments of the orchards. We're looking at tree foliage mapping, and also working with other companies to bring in other data sets and really build out a much bigger picture about what's happening in the orchard. Our dream for this is that in not many years time. A grower can wake up in the morning. Look at his iPad. And see what the recommendation for spraying is for today. Based on drone flights that have already happened. With drones self launching out of the box earlier in the morning. You can look at the ones that he wants and say. Go. And robots will self launch and go and spray the orchards. And the ones that he's not sure about, he can send that data across to his agronomist who can look on their pad and say, yes, no, maybe, maybe even drill into specific images, look at the high resolution data, try and work out why these recommendations been made, and they adjust it as they see fit. We're never going to replace the grower in the orchard. They know more about their orchards than we ever will. Hopefully, we can give them a good set of tools here. We can really help them to maximize their objectives from their orchards. Well, this is it. And I think I've had a similar conversation with Kit Franklin, who is one of the team at Handsfree Farm. The way that he puts it and obviously, there's a robotic solution around automation and much more about Arable. But I think he as a farmer and most of the team have always said they're not looking to try and remove the farmer or the grower from the equation. What they're trying to do is to use technologies to augment what those guys do and also to free up their time from laborious and boring tasks like sitting in a tractor, and give them more time to be working on insights to. Be thinking about sales, marketing, how to manage the teams better and to make best use of their exceptional talents and knowledge. Couldn't agree more. We really think growers have to be so many different people at once. They have so many hats they have to wear, they have to be a plant scientist, have to be an employer that has to be an entrepreneur, they have to be a negotiator as well as the landowner and all these other things. We want to try and make some of those tasks easier, take some of those tasks, off of them if we can, but also without making them need to be an international businessman or a technological visionary in order to make that happen. So it's about user operability. Most of the growers I work with are incredibly smart people. Never play poke with them, they'll take all your money. But they might not necessarily be engineers as well.

They might not necessarily. They have large thumbs : 28:19

-). We already expect farmers and growers to be so many different things. Expecting them to be data scientists, engineers as well is unreasonable. So obviously you mentioned the funding. I'm assuming therefore, if there's any VCs listening, you'd be interested to have a conversation within our more extensive kind of audience and member network. Are there any other people that would be useful for you to speak to, you'd be excited to speak to for collaborations or anything else? Definitely the two that spring to mind immediately. Obviously any growers or any fruit sellers that are interested in talking more about how we can make this work with you. It's always great to have a chat, even if it's a, no, that this isn't something you want. But so far we've had really positive response. Do give us a call or get in touch with us through LinkedIn or email, but also anyone, as you say, who thinks they might want to collaborate on this. In particular precision intervention systems in Orchard or other data platforms. We are particularly precious about this. We don't think that just one player is going to win the space. There are a few out there who are being very tight lipped about their data. We're quite happy to share, to collaborate and integrate because we think that's how we get the best solutions to the grower and that's the most important thing from the very top. So anyone, especially with the precision spraying systems who wants to use more of Outfield data to help their growers get more out of what they're doing, I love to have a chat. Amazing. Well, Jim, thank you so much for taking the time to have a chat. It's been really interesting, especially to get some updates on kind of the latest stuff that you've been doing and where you plan to go. I mean, is there anything else you'd like to say before we kind of wrap it up? Only that this is a fantastic space to be working in. It's a real pleasure to be working with growers. It's incredibly exciting at the moment. And I guess for any young people who might be listening, do consider agriculture as a viable career path now. It's not just living behind the steering wheel of a tractor anymore. There are incredible things happening with machine learning, drone systems, AI data and saving the world. And so, yeah, if anyone's wants to reach out and talk more about that, always happy to. Absolutely second that. I think it's one of the most exciting things about technology, taking such centres stage in agriculture and food. It's really opened up and changed the cultural perceptions. Jim, thank you. Bye.