Erik: Welcome to the Industrial IoT Spotlight, your number one spot for insight from industrial IoT thought leaders who are transforming businesses today with your host, Erik Walenza.

Welcome back to the Industrial IoT Spotlight podcast. I'm your host, Erik Walenza, CEO of IoT ONE, the consultancy that specializes in supporting digital transformation of operations and businesses. Our guest today is Don Miller, Chief Sales and Marketing Officer at Globe Tracker. Globe Tracker specializes in supply chain visibility telematics and sensor data solutions, enabling improved performance and customer retention as well as new revenue streams in key supply chain verticals.

In this talk, we discussed the digitalization of cold chain freight, looking at the vessel, the reefer, and at reefer generators. We also explored how ROI can be accurately calculated for a complex IoT solution to guide the sales decision process.

If you find these conversations valuable, please leave us a comment and a five-star review. And if you'd like to share your company's story or recommend a speaker, please email us at team@IoTone.com. Finally, if you have an IoT research strategy or training initiative that you'd like to discuss, you can email me directly at erik.walenza@IoTone.com. Thank you. Don, thank you so much for joining us today.

Don: Thank you. It's great to be here. Thank you, Erik.

Erik: Don, I like this topic because it's a very focused company, and so this really will let us understand a bit more the needs of the industry and technology adoption trends. And I guess here we're talking about seagoing freight industry. But before we get into the company and the industry, would be very interested in understanding where you're coming from. So I know you've been running your own company, Don Miller Consulting Services since 2007. And you have a background also in TrakAssure and other tech in this area. How long have you been in the market today?

Don: I've been in the market just around 12 years now. We've been focused primarily on IoT devices and global deployments in and around cold chain. So, current engagement with Globe Tracker is really focused on end-to-end supply chain visibility around cold chain. So we deploy technology that allows us to track monitor even down to the cargo level from manufacturer point right to end destination.

Erik: So, the company is really focused on cold chain and is this primarily, I said earlier, it's ocean going freight, maybe I was wrong there. Is that correct or are you also covering land?

Don: We do cover land but where we've cut our teeth, so to speak, has been on the refrigerated container market segment which is cooling containers that the ocean lines own or lease. And we outfit their fleets with an IoT device. That IoT device has multiple radios, cellular, Bluetooth low energy and LoRa, and those gateways either local to talk to sensors inside the cargo space, other tags in and around the reefer container itself. And then we have a full interface to the reefer machinery with protocols given to us by the reefer manufacturers, the ones that make the cooling unit that would be a carrier Transicold or [inaudible 03:55] or Thermo King.

Erik: I want to go into some of the industry terminology here, just to say I'm not from this. But maybe first cold chain, I guess it's a kind of a wide variety of things, but is it typically like food products, medicines, what would be typically moved here?

Don: Exactly, I think what you see in the market is 40% of the refrigerated container market moves perishable goods, whether that be fruits and vegetables, or meats, and things like that, and even pharmaceuticals that have degrees of shelf life, active pharmaceuticals as opposed to just shipping pills and other things.

Erik: And then you mentioned reefer, so if I look on your website, you have three solutions. You have the Vessel, the Reefer, and Genset. The vessel, I can understand, although I'm interested to understand a little bit more what you're deploying on the vessel. Can you just define Reefer and Genset for us?

Don: Sure. So Reefer is terminal is a term used in logistics to delineate a piece of equipment called a refrigerated container or a refrigerated asset. Because over the road, refrigerated trailers are also called Reefers. So that's where you get that term. Genset is a shortened version for generator set, which is usually a diesel powered generator that powers the refrigerated container or Reefer when it is either on a train or maybe even on a barge or sometimes over the road on a truck.

Erik: And then the reefer would be used multimodal, so you would take it off the bus, put it up the truck onto a ship and so forth?

Don: If we go back years ago, the refrigerated container segment was pretty much port to port, and then they would bring it to an offload distribution center and then move all the cargo into a refrigerated trailer. But these days, the containers increasingly are going inland so that all the cargo is placed inside the refrigerated container just once and then goes right to final destination.

Erik: And then the market would be divided between companies that, I guess you have the large brands that operate their own Reefers, and then you have a lot of logistics companies that offer this as a service. But what does it looked like from let's say we could consider the end customer or client who is moving goods, and then there's the different service providers that would be leasing space on this, and who would be the one that's actually your customer?

Don: So, in the current segment, the shipping line is our customer because what we do is we provide data to the shipping line on their refrigerated container fleets. We can pull alarms in near real time. We provide a system on the vessel that allows them to remotely monitor all of the refrigerated containers on a vessel above and below deck remotely, so they don't have to go out in inclement weather and check these things and constantly be writing things on spreadsheets: everything is now collected and sent to the cloud and they can process all that data.

The second piece of it that we've implemented using the local area networks that capability we put in the units would be to leverage technology called LoRa, and that technology allows us to communicate from our device which is on the outside of the refrigerated container into the cargo space into pallets and boxes and on the walls and such. So we can put sensors such as temperature, humidity, and even door sensing inside the container, giving an extra layer of both security but also visibility for the shippers which are the ones who own that cargo, or even the 3PL or the logistics company that's transporting the cargo on behalf of the shipper.

So we're talking about two separate market approaches, the permanent install of the IoT device for the shipping line and then subsequent wireless sensors that address shipper market or 3PL market.

Erik: And just so I understand the value proposition, so for the shipper, the logistics company that's being paid to move the goods, this is kind of an asset tracking trace and asset monitoring solution. And I guess they have SLAs that they need to meet, so this helps them to monitor those. So I guess you could be doing different things here. But you're tracking location. What else are you tracking?

Don: So, traditionally, we do track location. We also track G forces, so shock events. We track door activity. So there's been a lot of insertion of illicit goods around the world. You may have seen stories out of Latin America with bananas being filled with drugs and things like that. These are some of the typical things that happen in some of the trade lanes and they'd like to put a stop to it.

So in order to do that, we verify from the door sensor, when the door was closed and what the location was, whether it's in the load the stole load location, so we know where and when the door was closed. And then if anywhere between the time it was closed to the time it gets on the vessel and even to the end destination, if that door was opened, and it wasn't supposed to be, we generate alerts on that.

I think what you'll start to see is that Customs and Border Protection will become increasingly interested in solutions like this because it's not like a traditional locking device that's on the outside that people can cut like a bolt sealer, even another type of device that they can just rip off or try to throw away as they're doing what they do. This is actually inside the container. So it's a hard to spot and certainly hard to access without actually breaching the container.

Erik: Okay, interesting. The data in this case is owned by your customers, the shipping company, does anybody else have access, or how does access to work here?

Don: So increasingly, the shipping lines have started to share data with their customers. One of our biggest customers and is Hapag-Lloyd; they're number five in the world in terms of volume and size. And they have been at least introducing that product and the visibility to their customers on a platform called Hapag-Lloyd Live. So they offer this visibility service to their customers. I'm not sure on the commercials on that, that's totally with them. But they do have the data and they are sharing it.

Now, the data from the machine itself, they probably don't share everything that comes from the machine because it's probably not relevant to the customer, but compliance of temperature is: so supplier, return air, and even potentially alarm information could be sent to the customer so they would know whether their cargo is at risk at any point in time.

Similarly, so when we talk about wireless sensors now that go in the pallets and boxes, it's very similar to the data logging market. That's where people put in temperature sensors themselves because they weren't traditionally getting information from the shipping line, and they want to have also more granular temperature. They want to know exactly what the temperature of my pallet is inside the container.

And today, a traditional refrigerated container cannot do that. It can give you a supply, what temperature they supply the air at and a return which is what temperature comes back after it's passed through the entire Reefer space. So really, the optimal temperature is between those two numbers. When you have a sensor that's palate level, you get an exact reading.

Erik: You mentioned that regulators are getting more interested in this data as well. Do you know, is it common practice at all to give them access to the data feed or is that more if there's an issue that they might request access?

Don: I think as we've seen with all the slowdowns due to COVID and the trade imbalances and all the backlog, anything that can be done to speed up customs clearance is probably a good thing. I can tell you that an extra day in the terminal can cost a shipment almost $700 these days. So anytime you can expedite something getting cleared faster, you'd want to do that. I think these initiatives have been going for a long, long time.

And I think soon enough we'll get closer to more use of what we call smart containers in order to facilitate faster clearance times, because that along with initiatives in the US like CTPAT would allow a trusted shipper to load cargo from an external destination coming in and have it monitored the whole way. And if they were able to share the lock status data and other temperature data to ensure the load was compliant, there's no reason why it wouldn't need to be inspected at that level, if it was a CTPAT program, for example.

So that's like one example of something that could be facilitated and certainly save a lot of cost and a lot of time to both the shipper and the logistics company.

Erik: Interesting. So, I probably using the terminology wrong. So, logistics company is the company that's moving the freight. The shipper, if we're talking about let's say a brand, so maybe I'm thinking about this wrong, but I'm thinking about a dairy company or a pharmaceutical company as the shipper, would it be the same sensor suite, or is it a different tech stack that you'd be providing to them?

Don: It's a different tech stack. So they would be utilizing a small sensor, something like this, a wafer thin and that would record the temperature and they could put it anywhere inside the cargo. And basically, that set of data could go directly to them because they own that data. They're buying the sensor and they're putting it in the load.

What the shipping line would facilitate is being the network partner, because their containers are now outfitted with these devices, and it makes them run in a sense their own network. And these sensors can roam on their network. And it drives value to the line to have that capability because now you can take this sensor which is the similar price of a data logger. But anytime it's connected to the lines infrastructure, it becomes a real time device for the same price. So there's a big value proposition there for the shipper to use these things and they're now available in the market.

Erik: So if I'm looking at your website, is that the Sense Vessel would be what you're selling to the operators?

Don: Sense Vessel is an additional solution that allows full visibility at sea. So if I was to take the hierarchy, you have the sensor here that communicates to the core device in the reefer and that core device in the reefer communicates to the gateway station on the vessel which is called GT Sense Vessel. And then from there, that data could be kept locally or can be sent live via backhaul satellite to the customer and shipper. So that means that you could get near real-time information, even in the middle of the ocean sent back to you on the status of your cargo, which is really not been something that people have been doing for very long.

Maersk Line does this today using cellular base stations on the vessel. We work with a company called WMS, and they assisted us with a solution that goes and uses 4G LTE to collect all the information from the containers above and below deck, including the sensors.

Erik: And I think he  reached out to me, actually, because you’ve listened to our D-Orbit podcast, which is around satellite connectivity for IoT devices. What do you see there in terms of improvements in connectivity or cost structure? Have we already started to see some change, or is this maybe in the years to come?

Don: Well, I think it's still in development. But it looks very, very promising that you could have an application where a device went directly to satellite and didn't have to go through the cellular networks as traditionally listed. I think pricing will always be challenging, because the logistics space they've recently started to make money. But this will end up all evening out at some point the next year or so. And then once it does, then the pricing and the costs will come back down, and everyone will be back to a little bit more of a reality that everything is back to where it was, which are tight margins for the logistics space. And when that exists, then challenges around outfitting assets, whether it's containers and/or other logistics equipment remain.

The interesting part about the director satellite is cellular infrastructure today is very heavy in the bigger cities. But many of the countries, when it comes to inland or across land, transportation, the cell signals are not very strong. So whenever that exists, I always see opportunities for direct satellite feeds. Because if you talk about mining operations in the north, if you talk about monitoring a pipeline or something like that that's going through vast amounts of wilderness where there's absolutely nobody there, I think there's some really key applications for that. And logistics is certainly one of them.

Erik: Yeah, absolutely, it's going to be really interesting to see how this technology evolves. It sounds like even when I'm talking to the experts in the market, they're still exploring. I think there’s guy from D-Orbit mentioned that one of the Chinese providers of lower Earth satellites was selling images as NFTs. So he was basically trying to figure out how do I monetize this data that I have, and he hadn't quite figured it out. So selling unique images for holidays seemed like the quickest way to make some money.

Don, a little bit of a side question here. But when I'm looking at your Sense Vessel solution, you have something mentioned here in one of the bullet points, “Single solution installed on Bridge and Monkey Island with minimal vessel infrastructure.” What is Monkey Island here? This is a pretty unique term.

Don: So, right in front of the bridge, there's a station area where they usually set up all the antennas and/or satellites. And that area is called the Monkey Island. And if you look at the traditional pirate vessel, you would have seen these long poles with someone in there in what looks like a cage up at the top of the pole of the ship; one of the large mass, you'll see someone standing there and they were meant to be a lookout. So these are typically high places on the ship where you can look out from.

And our solution doesn't involve any drilling, so we don't put any holes into the poles or anything like that. We do have to run some cable conduit to access the V-Sat connection or the satellite connection on the vessel. But in general, the install, we've and came up with a mobile version of it. So that if you had a lease vessel and you only wanted to have the vessel for say six months or a year, you could lease a vessel with the capability and then take the system off and move it to the next vessel. So we've made it so that one set of antennas, it's two antennas, along with a network in a box type of thing, all set up, probably in just hours really, not days but certainly hours, maybe five or six ours, you can be up and running with your own network, and it monitors all the containers that have the devices.

Erik: So I think we've covered two of the products. And then we have this third which is Sense Scan Set, which you mentioned is deployed on the generator. Is this more of a kind of an asset monitoring so predictive maintenance, maybe location, fuel consumption, or what's the [inaudible 20:26]?

Don: Depending on what comes from the generator set itself, some of them have a controller, which is a mini computer that provides and generates data and others are analog, meaning that they only have limited parameters. So on a limited parameter generator set, we would take battery voltage. So understanding when the generators turned on and off by the fluctuations in the battery, we would also put a fuel probe in there and monitor the fuel level. So we would monitor the fuel level when the generator set was on or even when it's turned off. And you can set thresholds for loss of fuel.

So if someone is stealing your fuel and empties the tank by X amount, you'll get an alert generated. And of course, we do provide location, so GPS location of where and when these alerts occur, and of course, G-Force or impact events, if these generator sets get damaged, because there are thousands of dollars. So these are part of what we use to help them along the cold chain journey making sure number one, they have enough fuel in the Genset when they need one, they know where they all are when they need them, they know if any of them have alarms, and then we can set maintenance interval based on engine runtime. And that allows them to stay on top of their equipment remotely without having to go out and run inspections all the time, they're just getting all the data live.

Erik: And these are also going to be typically owned by the operator, is that the case?

Don: Parading it on behalf of the line and supplying them with a service where they take care of the supply of equipment. So it's a little bit in both cases there, where you'll have someone supplying a line by offering the equipment and managing it, and someone just simply leasing it to the line or the line owning it themselves. So there's a mixture of things.

But they're critical to keep refrigerated containers powered when they go inland. You might see a generator set attached to a chassis, it's called an Under Mount. And they are sometimes there's a cage built in under the chassis that sits there where they have the generator set in there so that whenever they put a refrigerated container on the chassis, they can quickly plug it in.

Erik: Okay, clear. So I'm curious to look a little bit more on the market side here. This industry to me feels like kind of a global top down industry because you're moving things around, you're moving a reefer from X to Y to Z, as opposed to maybe some other industries where somebody manages a factory, the factory is in this city, and they have a lot of maybe autonomy over that factory. From a customer perspective, where in the organization are you selling is? Are there local regions where you say I sell this to the Singapore team? Or is it more of a global offering and you're selling to a particular unit within a company?

Don: So, there's a couple of different segments in our business. But if I was to look at a large line, like a Hapag-Lloyd, so they’re a global line. So the decision from that standpoint would be made in their head office, which is Hamburg, Germany. So that's where you would go to put your offering across and then go through the testing and such whatever they do to make sure that you're the right technology for them.

If you're talking about individual regions, there are many, many regional lines. So Seaboard Marine out of Miami is one of our regional lines. Dole Fruit out of Costa Rica is also one of our regional lines, and they can sometimes run from Latin America to Europe or Latin America to USA but they're not fully global with their set of equipment that they run on their own. They do sometimes carriage out their cargo with the bigger lines that take it everywhere else around the world. But for their own equipment, they stay local.

And with Seaboard Marine, they run into Mexico Caribbean, around the USA, Houston, Louisiana, and head up the coast and even up the East Coast as well, and then down to areas like Costa Rica or the Caribbean. And in those cases, that's more of a regional line. The advantage of a regional liner, they're busier because the trips are shorter so the equipment turns more often. In other words, a large line may look to get 4.5 turns on their fleet per year, so that means 4.5 uses in one calendar year. Whereas a regional line may get between 8 and 12 uses of the box in a calendar year, and that changes the dynamic on the pricing and their profitability point because they're using the service a lot more.

Erik: That changes the pricing that you offer them, well, we haven't gotten into the pricing model yet, I'd like to…

Don: No, because typically the regional lines are much smaller in terms of size of fleet, because they're so concentrated and their equipment is moving so quickly, they can make real benefit of the use of it. So usually, when we price we price based on volume, and that's the price difference.

Erik: So also on the market side, you have a solution where, at least for me, it feels like the data has value to different stakeholders, different functions, maintenance operations, etc. So who would be the key decision makers or where would the budget come from? Is it coming from different buckets? Is it coming from a centralized bucket? Or is there basically one central function that would typically be purchasing this solution?

Don: I think in the sales journey, what's typically happened is, in order to sell a system like this, you have to go to the people who are utilizing the asset, and that would be the maintenance and repair team. And of course, you want to go to equipment management, which is in Hapag-Lloyd, it would be called container steering. And so you're grouping in maintenance and repair, those who are repairing and maintaining the equipment with those who are responsible for the movement of equipment in general. And those two come together, and you put your value proposition there.

Now, once you start getting into the sensors that are in cargo, and you're talking about commercial products, so now you bring the commercial teams in to have a look at the offering and see where this might fit within their group set. And can they take advantage of that? Is there a differentiation here? Are there special customers that would really love to have door sensing for some of their lanes because they've been subjected to a frequent breaches of an entries of illicit cargo, things like that?

So I think you do have to involve multiple tiers in the line, and the bigger the line, the more the tears. So that's just the way it goes. But this is definitely by far not an easy sell, because again, I think the refrigerated container manufacturers have now introduced some products that are pretty exciting for the market, those are called Smart PTI, and PTI is standing for Pre-Trip Inspection. And Pre-Trip Inspections are a necessity around the world before a reefer container goes on a trip, it runs a self-diagnostic. Well, someone has to go and run that diagnostic, and it could take you several hours to run it. And from that diagnostic, you get a result. Is something wrong with my container? Do I need to fix something before it goes on this trip?

What the manufacturers have come out with now is a diagnostic that runs all the time when the containers running. So now they can actually have that status anytime they want, and that means they can skip these inspections, which depending on where you are in the world can cost you anywhere from 25 to close to $100 per inspection. And again, if you go back to the four trips or 4.5 per year, and you do the math, you see that you can come now and really present an ROI solution based on even just one parameter.

Erik: And then the last angle to look at this would be from the distribution side. So is it director through some integrators or channels? And I think one thing I'm interested here is, what my understanding of your offering as you're selling the sensor, so the ability to collect data, to move data, and also a platform to visualize or access the data, my understanding is you're not selling algorithms that might be running on the data. So there might be other value added services that other service providers would be providing around the data sets that you have access to it. Are you typically doing a direct sale, or are people repackaging your solutions together with other technologies as some value added solution?

Don: So I think because the technology has really just started to flourish in the last few years, everyone's been playing a bit of catch up with just the difficulties of deploying a solution, where today we have almost 100,000 units scattered across 140 countries, and the challenges of that.

But put that aside. Fleet Management, how you manage the assets, how you know that the devices are communicating, how you know a healthier reefer is, how you know where they are, any alarms they have, there's a ton of applications that can be layered on top in terms of analytics of these of these devices, and the fleet in general, because you are collecting a lot of data.

In this market we're in now, I think hardware is the necessary evil. You need to actually get the hardware to start extracting the data. If you look at the ocean visibility segment right now, much of it is run on EDI messaging, which is a messaging system that's been around for 40 years, and it's based on different points of data, none of it is live. So when a container comes off a ship, an EDI messages generated. When a container leaves the terminal, and gates out, there's a message. And some are real time and some take days, and some, in some cases take weeks to process into the system.

So, now that you can put devices that are near real time on a fleet, you can actually accelerate these messages, not only for all the location points that you used to getting, but also diagnostics on the machine itself. And I think the combination of those two things makes this extremely attractive for visibility operators who are working on behalf of the shippers. I still think that there's a data ownership question to be worked out between the shipping lines and the shippers themselves and these third party visibility platforms. But access to the data is going to be key.

Erik: Yeah, that's the thing everybody's trying to figure out right now. We’re dealing with a case right now looking at machine learning in manufacturing, basically, analytics in manufacturing. And it's the same issue of until you have the infrastructure to collect in and process that data, it's really hard to make the business case because the first business case has to justify the investment in that infrastructure. So everybody's trying to basically figure out how to justify that investment. But of course, once you have that, then adding use case number 10 is pretty easy because you already have the infrastructure in place.

Don: Well, if you think of where we're headed, we've got 100,000 units installed out running around every day. Whenever they're powered, they're reporting every 10 minutes. So if you just think about that alone in terms of data points, you're talking about millions and millions of data points every single day. And so you can get really, really solid data on operations of a fleet very quickly, doesn't take long to get it. But in order to get it, you have to be the one that installed the device to even understand how to mine it, otherwise, you're never going to get access, because the lines are the ones who own that fleet.

Now, the sensor that I spoke about it's new to the market: we're the first to develop it. And I think what's interesting about that is you're leveraging the infrastructure on the line, and they become your network partner in delivering real time data for you that would otherwise simply be stored data that you would retrieve later. But now you have an opportunity to get it real time. So I think it's exciting. And I also see the technology evolving to the point that very soon you won't actually need this: there'll be infrastructure just with some of the technology we spoke about earlier, where you could have a sensor that goes direct to satellite, and then you don't need anyone else’s permission if it's your cargo.

Erik: Very, very interesting. And then the business model here, so you mentioned that it's volume based, is this a subscription model, is it a capex model?

Don: I think we've had to be versatile. In our field right now, 70% of all the containers are leased. So with that said, they don't own these units. There are lines who love to do the capex upfront and pay for it because they want to own the equipment, they want to own the telematics device. And there's others that will go with the leasing model where they won't have anything upfront, and they will just lease the service over a period of time along with warranty and everything else. So I think it does vary. But typically, the model has been hardware upfront, plus a monthly service cost.

Erik: So I want to get into a couple of case studies. But before we go there, just if we look forward a little bit, what are the technology development trends, the industry trends that are most exciting to you? Let's say, if we put a horizon of five years on this, what do you see coming out in the next five years?

Don: Number one is I think that narrowband IoT, Khadem 1, LoRa will all be in a pretty big battle for market share in the logistics space. Both technology is pretty exciting, offer pretty decent value, and I expect that there'll be a lot of adoption in that realm. With that said too, what I also see is that the price points for tracking systems will actually continue to come down.

Once we get through this chip shortage, there's a global shortage of chips, so the prices are a bit stable now and a little bit inflated from the shortage. But once that passes, I expect to see lower cost devices that have longer life and such and could be deployed on larger segments. We think dry containers, certainly something that's coming of age now and that in the next year or so there'll start to be some exciting developments and in that area, and you can already see it starting to appear. But I think it'll be even more clear over that time.

There's no doubt in my mind that we're going to get to a point where all container assets in the world are eventually tracked, or at least most of them. There are certainly some applications that will never require it, and when the equipment gets to a certain age, it won't be needed. But for any equipment, 10 years old or less, I expect this to be a standard over the next five years.

Erik: Okay, great. Well, let's go through one or two cases. And it'd be great to understand the end-to-end, who, of course, you can anonymize the name, but basically, where is the customer coming from? What are their pain points? Walk us through the deployment. And it would also be really interesting to understand what are the challenges across a typical deployment.

Don: I'll take one in the pharmaceutical and one in fresh produce. So the fresh produce one, the farms are in Morocco, and the customer is in UAE, and you're shipping blueberries that are now produced in Morocco, going to UAE. Blueberries are very temperature sensitive product. They're put in water-controlled atmosphere containers, and they're sensitive to deviations. And this is one of the cases where they're loaded right at the farm, and they're delivered right to the end customer. And they have to go through the med, and then cross and so on.

Again, monitoring temperature, but also gases, CO2 and CO2 as part of the solution in order to maintain freshness and shelf life are some of the things that have been going on for quite some time. But I think we're actually getting better at it. The shipping lines are improving their skills. The machines are improving their capabilities. And we're just seeing more and more of that.
 

If I take another use case around blood plasma, very expensive, life critical medicine, and obviously, this is one of the things: blood plasma saves lives, but it needs to be stored at minus 45. So you're talking about a shipping container that transports the plasma from one hospital, or one health facility to an end user hospital facility, and it could cross the Atlantic. So it could be a location in Europe from a location in the United States, for example, and you're talking about a journey that would take you close to two weeks. So two weeks of storing that in order to make sure the temperature is at minus 45 and doesn't drop below, because minus 20, you could spoil the plasma.

So there's a lot of these types of applications that exist out there. And sure, you could call them niche in a way, but I think that more and more everyone's trying new things. We're seeing avocado shipments from Chile to China now, because they've started to embrace the taste of avocado over in Asia. And I think this is also a fruit that's really, really sensitive and requires controlled atmosphere.

So there's many applications around how you're going to pull off an end-to-end and keeping it at optimum temperature and optimum gas levels for that period of time is a challenge. Because as you probably would guess, when you're moving a container on land, it needs to be powered by a Genset; when you move it into the terminal, it needs to get plugged in; when you move it from the terminal to the vessel, it needs to be unplugged, and then plug back in when it gets on the ship. And the same thing on the other side of the supply chain, all while maintaining the temp.

So there are challenges for sure, and these are some of the things that get looked at. Part of what our value proposition is, is that when these periods, we call them unplugged periods occur, you can set a threshold for how long a container can be unplugged before you want an alert. So we do that to assist the lines and making sure that they're not missing catching a cargo incident that cost a lot of money or lose a shipment.

Erik: So one of the most common challenges that I've seen around IoT solutions is the challenge of defining the ROI. And I guess in this case, I could see on the one end defining it around reducing scrap rates. So if we have alerts, may be set more optimally, we can reduce scrap and you could try to calculate an ROI based on that. Or you could look at it more from an operational perspective to say this will increase efficiencies for maintenance and operators and save labor in different areas.

So I guess two parts of this question, one would be how do you do this? What are the metrics that you would typically use? And the second would be how confident, is this done more at a qualitative level or can you really give a fairly quantitative dollar figure and say, based on your cost structure we can save you $500,000 per year with this investment?

Don: Yeah. I think, Globe Tracker has been successful, because we do have an ROI model. And we worked on it with close to 20 shipping lines. And we use to two or three reefer equipment manufacturers also weighed in on the solution, and the model. So what we allow a customer to do is we put in a set of industry standard costs, what it costs to repair a container, what it costs to move a container, what it costs to run a PTA, what it costs to run a software upload, meaning change the software on the container, because as the manufacturers release new software, they come with enhanced features, maybe power savings, other things. You need to upgrade a whole fleet. How do you do that? And what does that cost?

So you enter all these costs and then you give a column where the customer can actually work with their team internally and validate those costs. So they can change those costs. And each line operates a little bit differently. And they outsource some work. They internally have work here, not there. And either way, they've got to come up with a model.

And what we found is, after going through our model, every single line was realizing a positive ROI. And they were breaking even on the solution some as early as two months, but none over 18 months. So once you've reached that and you've had enough people validate, you know you have something solid.

So I think we have a list of parameters. Some require a process change and some do not. And I think the most valuable ones are the ones that are just there and don't require a process change. Because if they don't do anything new today, except get this data, they're still going to have to act on what they get. That's a given with anything IoT. You're going to get information, what are you going to do with it? And it's information that you didn't have before. So it does require something special. But at the end of the day, having them change an entire process to just reap one level of ROI is difficult. But some are willing to do that because they can see the value in it.

Just like understanding shock, for example. So, if a container gets damaged today and you don't have device on the container, the shipping lines, they own this box, or they're responsible for returning that box in good condition to a leasing company at the end of their lease. They don't really care how that happens. The leasing company just says send it back to me, but it's got to be in the shape it was. So they've got to fix these things.

And when you get a shock, and you don't know where it happened, the shipping line pays. But the minute you have a shock sensor, and it gets triggered by a heavy impact event, that GPS location along with the shock level and where the shock occurred, whether it's a right to left hit, an up and down hit, that kind of thing will all be sent to the line. And then they can now assign blame for that. They may pay it. They may not. They may assign it to the terminal to pay it. But at the end of the day, now they've figured out something tangible that they said this wasn't us.

And I think that's what happens when you're shedding light on the supply chain. You'll find out the things that you didn't know where there are there. Just like in the old days, the lines didn't want to tell you when they would trans-ship stuff. And the transshipment is when you stop it at another location for a couple of days to wait for a second schedule on a vessel to come pick it up.

Most people assume they booking freight it's going to be A-B direct. Well, that's not really the case, depending on where it's shipping from and to. Sometimes you have to stop in Jamaica, sometimes you have to stop in Panama, and such and so forth. So what that has never always been made clear. But with tracking devices now they can see where these Transshipment spots occur. And if they weren't aware, they're certainly aware, and they can actually see how long they spend in those spots.

And I think all of this stuff really opens up the whole idea of an ocean predictive ETA based on near real time data, which is I think, where the future is headed to. That's where we're going with all this. That's what would differentiate us from anything a project 44 [inaudible 44:35], or any of those guys have is that we have near real time data, including near live data on the ship to be able to say, yeah, we know how fast the ship’s going, you could get some of that by AIS, but we're more accurate than AIS. So there's a lot of good tech out there and this will help.

Erik: Yeah, interesting. No, there's always one of the more exciting parts of running an IoT business is figuring out the unknowns, so you know that there's value in this data and then you have some initial assumptions about what that is and you build a business case around that. But really there's a lot of questions that you don't really know to answer until you start collecting the data and playing around with it and getting opinions from different people.

Don: Clear algorithms that you can use in the accelerometer in our devices to detect when it's lifted off a vessel, so I can tell you the exact lift point and time. I can tell you when it's moved, whenever it's moved, where it's moved, how many times it moved in the terminal, when it’s gated out. I mean, all of those things used to take dozens of people to do what I just said. And when you outfit the container with technology, what's really great is that the infrastructure around it doesn't have to be as robust anymore.

Erik: Okay, super interesting, Don. I think we've covered a good bit of ground today. Anything that we haven't touched on yet that's important?

Don: No. I think we spoke a lot about IoT deployments. And I would just say that for those attempting to enter a space like this, there's a few considerations. Number one is, where are your customers shipping to because you're going to need to make sure that your device can communicate on the networks in those regions? Not just two of them out of three of them, but all of them reliably.

Number two is, what are the compliances that are required for the region you're going into, whether it's certification of the device, whether it's duties, taxes, shipping, what you need to do to get the device there to get installed and pay the least amount of money? You need to understand where the best places are to install based on your customer’s deployments.

Third thing is, how you going to pair the device? So IoT devices are great on their own, but how are you going to pair them to the asset? Because eventually, you're going to need to marry the asset ID with the device. And how do you do that in a cost effective way that's global? Today, we use a phone app to do it, but we also recognize that there's challenges around that, and we found them, where you're trying to install in a terminal.

And there's one in particular in Chile that's actually built in an open mine pit. So it's actually pretty located pretty deep in the terminal, and we weren't getting any cell signals. So as much as we tried to pair, we couldn't validate the pairing to the servers because the information wasn't getting through. So we had to actually physically move the install location in this case to be able to accommodate it, or in certain times, we'll put up cell towers in the location to make sure that we can get the signal out. And it all depends on the use case. So I think there's just a lot of things that need to be considered when you're doing global IoT deployments. It's not an easy oh, I'll get a device and I'll just send it. No, it doesn't work like that.

Erik: Yeah, great. Well, maybe that's a good place to wrap up. I mean, we're still on the first leg of this race, so lots of innovation and problems to address moving forward. But Don, really appreciate you taking the time to walk us through Globe Tracker and your solutions and wish you the best in continuing to grow the business.

Don: Thank you very much, and it was a pleasure. If I can help with any more information, just let me know.

Erik: Thanks for tuning into another edition of the IoT Spotlight podcast. If you find these conversations valuable, please leave us a comment and a five-star review. And if you'd like to share your company's story or recommend a speaker, please email us at team@IoTone.com. Finally, if you have an IoT research, strategy, or training initiative that you would like to discuss, you can email me directly at erik.walenza@IoTone.com. Thank you.