UTSKRIFT 🔗
WEBVTT
00:00:06.400 --> 00:00:16.399
In the commodity markets where billions of dollars move on crop forecast, the difference between profit and loss often comes down to who has the best data in the time.
00:00:17.039 --> 00:00:19.760
Welcome to Feed Duft and Focused or public health.
00:00:20.079 --> 00:00:25.440
Taking a look at the big issues affecting the lifestyle, poultry, grain, and animal feed industries.
00:00:25.839 --> 00:00:28.160
I'm your host, Terry Murphy.
00:00:28.480 --> 00:00:35.600
This episode of Feed Dusting Focused is sponsored by United Animal Health, a leader in animal health and nutrition.
00:00:35.840 --> 00:00:44.719
You can learn more about United Animal Health and how they're working to advance animal science worldwide by visiting their website at unitedanh.com.
00:00:45.600 --> 00:00:55.039
Joining us today to talk about what they're doing to rewrite the rules of market intelligence are Gabby Nesri and David Schiff of Sat Yield.
00:00:55.679 --> 00:01:02.320
Gabby, let's start by having you tell us a little bit about SAT Yield and the general services that your company provides.
00:01:02.640 --> 00:01:05.280
Sure, I'll start with a little bit of my background.
00:01:05.439 --> 00:01:10.799
So I'm a technologist, I've been in the IT automation space for many years.
00:01:10.959 --> 00:01:16.000
I previously co-founded IT Automation Company that was acquired in 2021.
00:01:16.239 --> 00:01:27.680
After that, I wanted to apply what I have learned for many years in automation, data, and decision supports to significantly bring my experience to the physical world problem.
00:01:27.920 --> 00:01:43.120
And agriculture stood out because it was one of the world largest and the most important industries, but many critical decisions still depend on information that can be delayed, highly aggregated, or based on historical assumptions.
00:01:43.280 --> 00:02:00.799
So together with my co-founders, who are a team of scientists, agronomists, engineers that are coming from remote sensing background, NAI, of course, we provide independent crop and physical supply intelligence with satellite imagery, as you said, weather and soil data.
00:02:00.959 --> 00:02:11.360
And most importantly, we have biophysical crop models to estimate planted area, yield production, and how those expectations change during the season.
00:02:11.599 --> 00:02:31.680
The purpose is not to really bring uh imagery and to uh extrapolate images from satellite imagery, is actually to help buyers, merchandisers, analysts, risk managers to understand exactly what's happening to the physical supply and whether they have need to act immediately or not.
00:02:31.919 --> 00:02:43.840
So, David, talk to us about how SAT yield's satellite-based approach, how it's fundamentally differs from traditional crop forecasting models by, say, USDA or other government agencies.
00:02:44.719 --> 00:02:54.080
Well, you know, I think one of the things that that kind of separates SAT
Load full transcript
yield apart from a lot of them is that you know, we're not really trying to predict what the USDA is doing.
00:02:54.159 --> 00:03:01.439
We're trying to give you a weekly insight as far as what the crop looks like based on that, on that it at any given time.
00:03:01.680 --> 00:03:07.280
And so when I do that, when I say that, we start looking at at corn and soybean models in in July.
00:03:07.520 --> 00:03:15.120
And from there, we're able to basically take a take make a prediction on what we think that the final estimate of the yield would should be.
00:03:15.280 --> 00:03:31.599
Uh, and this allows, you know, procurement folks, uh grain companies, traders, gather an edge and be able to look at the overall aspect of the crop and crop and determine its side size while it's growing.
00:03:31.680 --> 00:03:39.039
And it affects so many different things in the decisions that they're make as far as basis levels, transportation costs.
00:03:39.199 --> 00:03:46.879
Do I need to go in and and bring in crops from other areas of the, you know, outside of my radius?
00:03:47.039 --> 00:04:00.800
So if you're an ethanol company or a poultry company or a cattle company, all these factors as far as not just my localized area, but my regional draw and is and a national draw is affected by how with what what we're viewing.
00:04:01.039 --> 00:04:08.960
So, Gabby, can you walk us through a specific example of where SAT yield data translated into say an actionable plan for a client?
00:04:09.919 --> 00:04:11.199
Yes, absolutely.
00:04:11.439 --> 00:04:23.279
So I'll start with really kind of explaining maybe the concept of what we have actually built, which obviously allow us to explain also the results and the outcomes.
00:04:23.439 --> 00:04:29.759
So as I mentioned before, we are not just taking satellite imagery and explaining what's going on inside.
00:04:30.000 --> 00:04:44.720
We actually created a crop digital twin, a virtual representation of a crop that basically is being developed as much as it kind of looks like in the field, right into our computers.
00:04:45.199 --> 00:05:28.800
So when we combine this environmental information to the crop digital twin, we are able to simulate from farm levels like polygons and conditions in specific areas of interest, you know, for example, uh crop insurers, um, and lenders and banks who actually are willing to insure that farmer, and up to basically financial services guys who are very interested to know at the state and national level what's got what's going on right now during the season, what's going to be um presented or expected uh in acreage, in planted and harvested, what's going to be the yield toward the season.
00:05:28.879 --> 00:05:33.120
Sometimes we have a weather event or situations that require immediate response.
00:05:33.279 --> 00:05:57.759
So we will be able, using our crop simulation and the basically the feed from satellite, to help them to understand what's going on right now in the field,
what going to be the um results or the outcome of the last weather events, and how much is it going to influence the yields, uh, which then translate to risk, uh price, and any other decisions that they might need to do.
00:05:58.240 --> 00:06:07.759
So, in practical terms, then how does this level of granularity um create this, create the information advantage when it comes to commodity trading?
00:06:08.079 --> 00:06:08.720
David?
00:06:09.759 --> 00:06:12.959
So think about it this way.
00:06:13.120 --> 00:06:26.160
If you if you have information first and you're able to kind of get it get an edge on what you what the crop size is, but let's take it on on a national level.
00:06:26.399 --> 00:06:29.839
And so when you're looking at it on a national level, how big is the crop?
00:06:30.160 --> 00:06:32.959
Well, the USDA puts out a port report monthly.
00:06:33.040 --> 00:06:35.360
We're putting out a report weekly.
00:06:35.600 --> 00:06:46.959
Uh and within those weekly things, you're able to make decisions both on a trading side as well as a physical side, a production side, to buy side, and it's a trickle-down effect.
00:06:47.120 --> 00:06:58.240
It's not only procurement people, but if you're a farmer and you've got access to this information ahead of time, you'll know how much you know, you'll be able to tell if it's on a field level basis.
00:06:58.560 --> 00:07:02.560
Are you able to, you know, market more grain?
00:07:02.639 --> 00:07:04.079
Are you able to move it to bents?
00:07:04.319 --> 00:07:24.319
And and so when I look at this and I look at it as a whole, how you know it's the decision making and the and the ability to make it over the course of a season and have all that information at hand that is accurate and can predict an outcome that you're not waiting four months to find.
00:07:24.399 --> 00:07:32.639
You're you're seeing it live and in person today, and in the reports, I mean, are almost in real time that we're uh that we're projecting.
00:07:33.360 --> 00:07:45.839
So from a practical matter, it it's it's information and being able to use that information in a timely manner ahead of what the USDA or what the market is is seeing.
00:07:46.160 --> 00:07:50.800
How does your data continuously update and validate its predictions?
00:07:50.959 --> 00:07:58.000
What kind of mechanisms are in place to ensure that the data remains viable and reliable for decision making?
00:07:58.399 --> 00:07:58.639
Right.
00:07:58.800 --> 00:08:11.120
So the beauty of this model is actually not relying on any historical data or relying on boots on the ground or field surveys or phone calls to farmers, etc.
00:08:11.519 --> 00:08:18.720
It is an independent, um, unbiased model that basically simulates the reality.
00:08:18.959 --> 00:08:27.360
As long as we get the feed from satellite on a three to five days basis, we are able to see what's going on out there in the field.
00:08:27.439 --> 00:08:48.399
Again, like it could be a single field,
could be entire township, county level, etc., and be able to see exactly what's going on almost in real time in agricultural life of like a corn and soy a week is a significant uh time that can change the entire trajectory of the yield.
00:08:48.480 --> 00:08:58.799
And we are able to get those uh insights through satellite imagery and then feed the algorithms and the models are basically simulated all the time.
00:08:58.879 --> 00:09:14.240
So we get to see how the crop responds to weather conditions, to soil moisture, to any other events that could actually stress the crop, and by that to influence the yield, and we can measure and report on that weekly.
00:09:16.159 --> 00:09:26.080
So, what crop types, what geographical regions, um, you know, what's what's covered, and kind of what are your plans for possibly expanding uh in the future?
00:09:26.639 --> 00:09:27.039
Right.
00:09:27.120 --> 00:09:31.600
Um, so we are a very young company, we've been around since 2024.
00:09:31.840 --> 00:09:41.200
Uh, we've been able to uh detect and classify crop types such as corn, wheat, soy, canola in Canada, sugarcane.
00:09:41.600 --> 00:09:50.879
Um, we are now focusing on mainly uh building kind of a complete map of the major global uh production area.
00:09:51.039 --> 00:10:03.200
So uh, for example, uh in America we will focus on the corn belt, soy belt, and wheat belt, um, which we are now running uh as we speak in season.
00:10:03.360 --> 00:10:07.360
Uh the next season coming is Brazil and Argentina.
00:10:07.440 --> 00:10:15.919
So we are planning to start with soybean and then move to the second crop safrina um along with the Argentina in parallel.
00:10:16.080 --> 00:10:18.240
And then the cycle continue and return.
00:10:18.480 --> 00:10:22.080
Now we have plans to scale to more geographies.
00:10:22.240 --> 00:10:33.519
There's lots of interest, and and um we know also there is lots of uh going on in uh war zones that you can't you know um step there.
00:10:33.759 --> 00:10:47.600
So Ukraine, Russia, for example, uh even in remote places like in India to bring some information from there and to provide um accuracy results of what's actually going on with the supply side.
00:10:48.080 --> 00:10:55.679
So how do how do traders, how do they typically integrate the data into their existing trading workflows or risk management systems?
00:10:55.840 --> 00:10:56.879
Is that an easy process?
00:10:57.039 --> 00:11:01.279
Is that something that um you know that that happens on a regular basis?
00:11:02.000 --> 00:11:07.679
The the how part is actually generating and creating those data sets on a weekly basis.
00:11:07.919 --> 00:11:09.840
I think the rest is quite simple.
00:11:10.080 --> 00:11:20.000
So uh we are able to deliver those data sets, the raw data, aggregated data as well as insights to different types of means.
00:11:20.240 --> 00:11:31.519
Um, with basic stuff like APIs, web services, integrations, uh, even like now everyone talks about
MCP to connect directly to your cloud or ChatGPT.
00:11:31.759 --> 00:11:44.159
Uh, but even going back to traditional uh way to basically send files uh or provide the FTP so people can actually go and download uh those data sets on a weekly basis.
00:11:44.320 --> 00:12:04.399
And finally, we develop basically a visual layer that you can interact with the data yourself, like a chatbot, uh, which allows you to uh query the information, query the data, ask questions, and receive uh the results in the context of your question.
00:12:04.559 --> 00:12:12.559
For example, if you like to see uh how Satil compares to USDA, you could actually ask the agent to do that for you.
00:12:12.720 --> 00:12:30.080
If you want to run a simulation of a supply and demand scenario, you can actually ask, and the agent would actually bring Satil proprietary data along with uh NAS and other uh um you know reports that are coming from the official governments.
00:12:30.480 --> 00:12:48.240
One of the things that we have is is a dashboard as well that we've built, and being able to ask the agent, you can query it, you know, get live reports on on in-season reports for corn, soybeans, wheat in US, Brazil, uh, Argentina.
00:12:48.399 --> 00:12:51.279
You can also get weather, you can get SD things.
00:12:51.360 --> 00:13:00.720
And the the cool thing about it is from a trader standpoint, is you can query what you want and then have a notification to where it comes to you every morning at a certain time.
00:13:00.879 --> 00:13:08.159
So if you're getting up at 5:30 having your cup of coffee and you have a report delivered to you, it'll give you the morning report.
00:13:08.240 --> 00:13:10.960
It'll kind of tell you what's going on in the various different things.
00:13:11.120 --> 00:13:15.200
So, you know, it to me the the dashboard is user-friendly.
00:13:15.279 --> 00:13:26.240
If you're, you know, if you're a trader or if you're you're a hedger, you're working in the in the elevator or or wherever it is, you get this data, you get this information, it's there, it's easy to read.
00:13:26.399 --> 00:13:32.320
Uh, you just pop it up, and then then you can, you know, if you've got stuff going on during the day, you can ask it as well.
00:13:32.559 --> 00:13:37.120
So any final thoughts or key takeaways you want to leave our audience with here today, David?
00:13:37.200 --> 00:13:39.120
And I'll come over to you next, Gabby.
00:13:39.840 --> 00:13:55.360
You know, I I think when when when I think of SAT yield and I and I look at it, what what attracted me to it was, you know, it's it's it's really the ability to to look at things more on a granular, granular level.
00:13:55.519 --> 00:14:15.200
Uh I'll put my farmer hat on and tell you that, you know, with my family's operation, being able to walk out into a field and see one thing is is great, but being able to triangulate it around there and look at the whole field is even better.
00:14:15.360 --> 00:14:18.960
Uh because it gives me the ability to make marketing decisions.
00:14:19.039 --> 00:14:26.240
It gives us the ability to look at our assets and figure out if there's an opportunity to save on transportation.
00:14:26.320 --> 00:14:33.360
Is there an opportunity where we can sell our crop, store our crop, do different things?
00:14:33.519 --> 00:14:34.960
Is there a shortage around?
00:14:35.120 --> 00:14:42.240
Because and and so from a practical standpoint, you know, it's really the independence for for a buyer.
00:14:42.399 --> 00:14:51.440
Try and you know, I look at it and and being able to triangulate around assets, prepare for expect the best, prepare for the worst.
00:14:51.600 --> 00:15:00.559
So if if you know you know what is is going on, you're not going to the coffee shop hearing one thing, you're able to look at it as a whole around your assets.
00:15:00.639 --> 00:15:11.279
So I think that to me is is the biggest thing of being able to look at the assets, look at the information, and make sound economic decisions uh to manage your risk.
00:15:12.159 --> 00:15:14.799
Yeah, and I'll make it uh much shorter.
00:15:14.879 --> 00:15:32.480
So uh if anyone is looking for an independent view of physical crop supply that helps you know the industry to see meaningful changes earlier, I would say that uh SATTIL is actually here to uh bring this uh quite relatively news.
00:15:32.879 --> 00:15:40.080
This episode of Feedstuffs in Focus has been sponsored by United Animal Health, a leader in animal health and nutrition.
00:15:40.320 --> 00:15:49.679
You can learn more about United Animal Health and how they're working to advance animal science worldwide by visiting their website at UnitedANH.com.
00:15:50.399 --> 00:15:54.720
I'm Sarah Muirhead, and you've been listening to Feedstuffs in Focus.
00:15:55.360 --> 00:16:06.559
If you would like to hear more conversations about some of the big issues affecting the livestock, poultry, grain, and animal feed industries, subscribe to this podcast on your favorite podcast channel.
00:16:07.039 --> 00:16:11.039
Until next time, have a great day and thank you for listening.