During this week's episode, Erik interviewed Gil Dror, the Chief Technology Officer, and Sammy Kolt, the Chief Product Officer at SmartSense. SmartSense offers a solution that empowers organizations to leverage wireless sensors, digital technology, and advanced analytics, resulting in actionable insights based on remote sensing and artificial intelligence.

Throughout the discussion, we delved into the ways IoT facilitates workflow streamlining and introduces a novel approach to data-driven decision-making. Additionally, we explored how IoT is making a significant impact on the healthcare and food services industries, where it is utilized to enhance safety measures, ensure compliance with regulations, and maintain stringent quality control standards.

Listen and enjoy the show!

Key Questions:

● What solution stack does SmartSense utilize for its software and hardware?

● How do they ensure the longevity and efficiency of sensors and equipment?

● What enhancements have they implemented to enrich the user experience through their IoT solutions?

● What exciting developments can we expect in the realm of Smarter AI?

If you're curious to know more about our esteemed guests, Gil Dror and Sammy Kolt, you can find them on LinkedIn:

Gil Dror: https://www.linkedin.com/in/

Erik: Gil, Sammy, thanks so much for joining us on the podcast today.

Sammy: Thank you. Thank you, Erik, for having us.

Gil: Thanks for having us.

Erik: Great. Well, I'm looking forward to getting into this. Digi is one of the companies that I first came across when I started to really dive into this space, so it's been on my radar for a long time. But SmartSense is a bit more new to me, so I'm really looking forward to understanding your business. But before we go there, I'd love to learn a little bit more about the two of you. You've both joined, according to LinkedIn, at least in 2022. So you're relatively new to the business. So I guess maybe the question that I have is, first of all, what excited you about SmartSense and Digi? Why did you join? Then was there any reason around 2022? Is it that there's kind of a ramp up or something, or was that just a bit of chance there?

Sammy: Yeah, so I joined in early 2022 when one of our managers from previous companies joined Digi as president for SmartSense, where Digi decided to increase investment in SmartSense, to invest more, to put more power into the IoT solutions business unit. I came back after talking to Guy and saying, yeah, there's a great opportunity for me to join the area of IoT. In my entire career, I've been in an enterprise software but never been in a company that combined both hardware and software. I thought it's a great opportunity for me. IoT is a market that obviously is highly sought after. A lot of people are working in IoT, looking at IoT. That was an opportunity for me to make a shift in my career and join this great company.

Gil: For me, I started my career a very long time ago in the Israeli Air Force. My background was really around electronics. But then I spent the majority of my career, like Sammy, working in an enterprise software, specifically in healthcare. What excited me about Digi was that it really combines the best of both. It's cutting-edge IoT with enterprise software and targeting healthcare amongst others. So to Sammy's point, 2022 was kind of a pivot for SmartSense where they decided to really put the gas on trying to capture this market. So I'm really excited to be here.

Erik: Got you. And so SmartSense in the Digi portfolio. Is SmartSense kind of the solution provider vertical, and then Digi is more focused on selling pure technology solutions? How would I look at the relationship?

Sammy: Yeah, so you can look at SmartSense as a business unit within Digi. Digi, obviously, they design, the manufacturer. They support all of the hardware predominantly in the IoT world and specifically infrastructure technology. And yes, SmartSense is part of more of the solutions business unit within Digi. We are very, very vertically-focused, application-focused, whereas the rest of Digi is more of horizontal technology. So yes, SmartSense is a business unit within Digi. SmartSense was actually created by acquiring four different companies, combining them together and taking the best from all these companies, rebranding it into SmartSense in 2018. Now SmartSense is a business unit within Digi.

Erik: Why in 2022 make the decision to really invest more in ramping up the solutions business? Is this just an internal company timeline — that you integrate in 2018, it takes a little bit of time to get everything oriented and have a strategy, and then this just makes sense in terms of timing? Or was it more market-oriented in terms of you've seen a specific uptick where this was needed in the market at this point in time?

Sammy: Yeah, I think it's both. I think there were several things that you mentioned. The integration, yes, it takes a little bit of time to combine the companies, so it was time to rebrand and go to market as SmartSense. The other thing is, COVID happened. So that pushed a lot of regulations and a lot of visibility that was required — at least on the healthcare and pharmacy side — to monitor, to make sure that everything that consumers are either buying or receiving, as in the form of vaccines, is kept safe.

Overall, the market, yes, we hear more and more about retail establishments, and hospitals, and pharmacies that essentially are obligated to provide safe and high-quality products to their consumers. So there's more need for automation. Workforce — we all know how people talk about the lack of workforce and lack of people, so there's need for more automation. All these things combined together pushed to look at SmartSense as, let's ramp it up, and let's go to market with more investment.

Erik: Got it. Yeah, COVID was really terrible all around. But one of the few silver linings was that, for digital solutions, it really forced a lot of companies to take strategies that they've maybe had on paper for a while and actually put them into practice in terms of implementing data collection and automation solutions. So I'm sure it was a shot in the arm for the business.

Let's look a little bit first at the business from more of a customer perspective before we get too much into your tech stack. So if I look at the industries you're serving, it looks like two big verticals. Healthcare is one. Then there's greater retail and F&B space that you've broken down into retail grocery, food services, convenience stores, and supply chain monitoring. But I guess there's a thread that runs through all of those. Then you have K-12, which I put a couple of question marks around that. It really stuck out me as a bit curious. How did you define these industries as the starting point for the business?

Sammy: You mentioned health care and what we call food service, right? Under food service, you can actually put in the groceries, the C-stores, the restaurants, K to 12. I know you have a few questions on K to 12, so I might leave it for later. But yeah, so this is how we look at things. Really, when you start going down to the details on what solutions we provide, when you look at health care — which, again, on healthcare, it's the hospitals and pharmacies and labs — the underlying solution is very similar across all these verticals.

Within food service, what we do and what we help our customers with in grocery is similar to what we do in convenience stores and in restaurants. Obviously, they have their nuances in each one of these. But at the end of the day, our customers want to provide safe and high-quality product for their consumers. And that's what we're there to help them. So that's how we see our verticals. Obviously, we're constantly looking at what other kind of adjacencies and other areas we can go into with our solutions. But that's how we look at our verticals.

Erik: Okay. So safety, compliance, quality control. These verticals all emphasize those as critical parts of the business.

Sammy: Mm-hmm.

Erik: Okay. Then the solution that you're providing, it's a hardware, software solution. But then you are kind of a solution business, so I guess there's some degree of customization there. Then there's also regulatory compliance. It's important there. So hypothetically, there's also potentially some back end that could be quite important for acting on the data and so forth. So can you walk us through what that solution stack looks like? Then which elements are quite standardized, and what are the elements that are going to be more customized?

Gil: Yeah, sure. In general, we try not to customize but more so to configure. So that's a slightly different approach. We tried to build our software and the hardware in a way that it worked for all of these industries without any customization. We just make them smart enough so we can configure them to the individual client needs.

In terms of stack, one of the first, I'd say, big priorities for us was to be obviously reliable and secure. Our system is designed really from the ground up to be working almost in parallel to your IT infrastructure. It's all battery-powered. It's all cellular-based. We have a gateway typically sitting in the location, communicating with the different sensors and uploading everything to the cloud. There's a lot of technology that goes around making sure that if there's no connectivity or there's some interference, the data is still collected and still uploaded and processed. So from a stack perspective, we definitely have a lot of technology around the hardware side in making sure that the data is collected properly and then uploaded to the cloud. In terms of the cloud, obviously, we have a very robust API with very strong ingestion capabilities that's able to take these millions of records that are coming in, and process them and really generate the insights that the client needs.

Just to talk a little bit about the use cases and in what Sammy was saying, if you think about a typical use case — whether it's retail, or pharmacy, or a hospital — you could put a thermometer in a freezer and see if your vaccines are in the right temperature right now. But you would really need to put someone to look at that thermometer 24/7 to know if there was any fluctuation or to know if something really was impacting the quality of that product. That's really what our system is designed to do. It has built intelligence to understand, okay, this is what this particular cooling unit is supposed to do. These are the patterns of temperature. Therefore, we know when this product is at risk, and we will alert the client for that. So from a stack perspective, it's really end to end — all the way from the electronics that collects the telemetry, all the way back to the machine learning algorithms that process it and actually tell to the client.

Erik: Okay. Maybe we can take the stack one layer at a time. So if we start at the hardware, I guess in a lot of these cases, they're going to have a lot of sensors already in place. They might be disconnected dumb sensors, but they'll have some kind of temperature sensor maybe in the refrigerator already. So when you go into a case, are you typically recommending that they deploy new Digi hardware? Or are you going to say, "Okay, you already have 200 temperature sensors, and we're going to integrate those data streams. But then maybe we're going to deploy in another 50 sensors to fill some data gaps"? How do you look at that in terms of — do you work with the brownfield infrastructure that's in place, or does that cause issues in terms of standardization of the data and interpretation, and therefore you recommend more standardizing around a new set of hardware?

Gil: Yeah, it's a great question. It really depends. We support both. In some cases, for example, we had a client that said, "Hey, look, we already invested in these cutting-edge freezers. They'd come with temperature monitoring. We don't want to put another sensor there." But these other stores, they don't have anything. So in that case, we would go with a mixed approach where we will take the data from wherever this product that they have. Plus, we'll send them our own sensors. We're very happy to do that.

The question is really, we try to ask the question to the client and say, okay, these sensors that you have, are they a long-term strategy for you? Is the vendor still around? Are they still supportive? Is that something that you want to keep and you're comfortable with supporting for the next 5 to 10 years? If the answer is yes, then absolutely. If the answer is no, typically, the client will say, "Look, this is already obsolete." Why don't you put a new sensor in? It's already connected, and it's probably going to be cheaper. They'll get the maintenance through us. So it really depends. But we can support both.

Sammy: And just to add to that, Erik, most of the equipment that we see that already has sensor in it, the thought behind putting the sensor there was more so to monitor the equipment rather than monitoring the product that is in the equipment. The reason I say that is, our goal is always to monitor as close as possible to where the product is being kept. Which means, that a lot of the time, if you think about walking into a supermarket when looking at all these display cases, our sensors are going to be located where the consumers are reaching in with their hand and taking the product. Because that's where you want to monitor: as close to the product as possible. In these situations, if this equipment do have a sensor, they're most likely going to be more in the back and monitoring the condenser and the fans rather than where the product is.

It's interesting because these are the types of nuances that when I came over, I never thought of — a refrigerator is a refrigerator, right? You open the door. It's cooling. A product, and that's it. But there are so many nuances that when it comes to quality of product and safety of product, you need to get into that level of detail. But as Gil said, we are happy to entertain hybrid approaches but also our own equipment.

Erik: I'm curious on your thoughts on, let's say, the form factors of sensors and where we're going. Because I guess, if you're putting a sensor into a commercial refrigerator, you have one or two sensors in there. The commercial refrigerator costs $20,000 or something. So you don't worry too much about the cost of the sensor and so forth. But if you want to then start getting more granularity, and you want to say, "Well, okay, we want to have this in each truck. We want to have it maybe in a crate when fresh produce is being moved around. We want to make sure that it wasn't set out in the sun on a dock for an hour waiting for the truck to arrive, so there might be these gaps when it's not in a structured piece of equipment. Then you get into situations where you're saying, "Well, okay. But the sensor has to cost — the form factor has to make sense. The cost structure has to make sense if we're going to actually put sensors and have a more granular data."

Where are we today? What are the situations where it makes sense to put a sensor because the structure works? What are the situations where there's still gaps in the data, because maybe the cost of the sensor is just too expensive or the cost of the connectivity? Then if those are gaps, are you expecting that we're going to address those in three years, in five years, when we get down to a point where you can take a $4-sensor and connect it for $5 a year and then fill in all those dark spaces in the supply chain?

Gil: Yeah, that's a great question. Even today, we definitely offer a wide range of price points and sensor sizes exactly for that reason. Because there are different applications. You might need something the size of a sticker because you want to track a shipment all the way to, "Hey, I have this massive freezer. I need to have three or four sensors connected to it and all these calibrated and things like that. So we require a bigger footprint." The question sometimes is not around cost, though. When you think about a typical installation, you have to remember that at the end of the day, there's got to be some change management on the client side to handle whatever is coming from the temperature monitoring system.

There's a limit to the capacity of the staff to what they can handle. Let's say you have a store, and it's got 100 freezers. It's just not going to be possible for the staff to manage so many. And so typically, the client will make a decision in terms of, "Okay. These are the ones that are the most important. They're the highest risk. Those are the ones that we're going to focus on." And all the other ones, they're going to monitor to some degree of maybe not invest in real-time monitoring, or maybe they will put some cheaper sensors, or maybe less sensors or something like that. A lot of times, it's not necessarily just the cost of the equipment. I think in the grand scheme of things, if you look at the cost of product loss over time, it doesn't matter what the sensor is. It's going to be a fraction of the cost. But it's more around the other costs that are involved with really protecting that that come into play here.

Erik: Yeah, got you. Okay. Well, why don't we then look at the software here, because I guess that's where you get into the business processes more, and then this topic? My assumption is that you have a back-end that I assume works quite well, and is quite standardized in terms of ingesting the data and doing some processing and integration. But then you have this front-end interface to the users, where they have to then integrate your product into their business processes and make decisions around it. That's the part where, at least, in my experience, there tend to be more headaches or more challenges. Because then you're dealing with human dynamics and human interaction with the software. What does that look like? What is the expectation for the user interface? Is this primarily viewing data? Do you take in another step, especially maybe in healthcare, and start integrating into sending them compliance alerts or integrating into different business systems? What does that look like currently?

Sammy: From a software point of view, one thing that is important to understand, we have tens of thousands of users. But our users, majority of them are very casual users if you talk in the terms of software. Because the way they interface with our solution is really when there's either alerts or reports that they need to get into. Majority of them are casual users, so we had to build our software in a way that would be really, really simple, easy to orient themselves into the software. You can't sit through several days of training to be able to work with the software. By the way, when we say software, we put obviously both the web application, the interface on the web, but also mobile. So it has to be very, very simple, very easy to use. If you're a pharmacist, if you're a staff member in a grocery chain, you need to be up and running pretty quickly once we install the software, or if you're new to the company. That's one thing about the software.

In terms of other things that we do from a software perspective — we mentioned analytics, machine learning — we do have users that are admins that set up and keep the system running, add new sites, add new assets. They're the ones that also consume a lot of the reports, looking at the company at an aggregate, understanding any insights, understanding any challenges that they have with maybe some brand of equipment that they have within the company. Our software really caters for both the casual user, the admin of the solution but also for our installers. We've developed a program that helps to really, really shorten the time to get up and running. I mean, we can get up and running in minutes once we put our hardware in a site. So that's how we look at software. Then Gil, if you have anything to add more on the back-end, I'll let you talk about that.

Gil: Yeah, I would just add that if we think about our capabilities from a software perspective, there's really a few points. Sammy touched on a few of them, which are around the casual users getting alerts. That interface could either be, hey, I'm just going into the web app to get the alerts, or I'm getting them through SMS, or I'm getting them through a mobile notification. We try to meet the users where they're at.

We also have another aspect of the system which is process automation. A lot of processes that are happening within the healthcare or retail are still manual. It's a bunch of logs and papers. They go around, and they check boxes. Hey, I did this. I did that. We automate all of that, and we combine it with temperature monitoring. So you could have an alert that actually triggers a process. In that case, the end user will actually use our web app or our mobile app to execute a process. Again, that's really a game changer because we're moving away from either not doing it, and just putting in a rubber stamp and saying, "Yeah, I did it." They're actually capturing who did it, when they did it, capturing the evidence right there. At that point, we take advantage of the fact that they have a mobile device that has a camera, that can capture evidence right on the spot. It's all digitally transferred to the cloud and stored. So there's a lot of capabilities around that as well.

Finally, more around the FDA and the compliance, our users can go in and see their NIST certifications through the application. They can download a compliance report that they can hand over to an auditor that comes in. So again, that's another use case that our end users are using the application for.

Erik: Got you. Okay. So it sounds like there's maybe three big buckets of users. There's the system admins. There's the more maybe office, let's say, the supply chain manager who has to print out reports and look at scrap rates and things like this. Then there's the frontline people who have to respond to alerts and use maybe more of an application version because they're on the go and working while they use the app on the floor. If you look at who your buyer is, I guess we're looking at supply chain solutions, right? So it's probably not like a store manager. It's going to be more at a corporate level. Then is it typically more from an IT perspective? Is it a supply chain VP who would be a typical buyer? Who are the other really important decision-makers in that buying process?

Sammy: It's interesting because I think, overall, the technology industry has been going through a transformation, where in the past, every big technology purchase had to come from the CIO office. It was driven by the CIO office. They would typically run it to serve the business. What we see today is a shift to more of a business-driven rather than technology-driven purchases. Now it's more engagement with the business. Then more specifically, we used to mostly work with the safety, quality compliance teams. These are the ones that would reach out to us to purchase our solution. What we've done and what we've seen in the last two years or so is more shift towards more of a cross-functional purchase. Because the data that we can provide — I know we've been talking about temperature mostly. But there's other things that we monitor that can help our companies that can help other functions within the business to get value from our solution. Things like, again, in healthcare, we monitor pressure in C02 and 02. We monitor open, close for doors.

Let's take, as an example, a grocery chain. Yes, safety, quality, compliance, obviously, they have a stake in this. Facilities, they will have a stake in the solution. Because when we monitor the product, we also monitor the equipment that holds that product. They are the ones responsible to make sure the equipment is kept maintained, and they can benefit from the data that we provide and the insights that we provide. Loss prevention, organizations — again, there's this constant interface between loss prevention who are responsible for reporting and preventing shrink and losses within the company and the other function. So they have a stake in this because if they now start getting visibility into where there are challenges with either equipment loss or product loss, they get early signs and can stay ahead of what they report on.

IT, again, even though you'll mention that our solution is pretty IT-independent, meaning that we can come in; we don't need to connect to the network — again, I'm talking from a technology perspective — we don't need to connect to their network. We can be pretty independent of their network, but they're still responsible for the technology within the company. Obviously, they're on the table. Supply chain, as you mentioned, Erik. Again, bottom line is: our buyers can come from multiple areas within the business. Typically, what we see is more of — So wait. Your data, what you're telling me is that other functions can benefit from that data, which actually makes it better case to take it up to management to say, "Yeah, with one vendor, we can support multiple functions within the company."

Erik: Yeah, that makes a lot of sense. I think you can look at a business case for an IoT solution as kind of a stack, right? You can get 30% of the value from this function and 20% from that function, and 20%. If you're able to serve them all, you can actually have a great ROI. But if you're only serving one, it might be a bit borderline, whether it makes sense.

You can also imagine cases. Let's say, if we look at healthcare supply chain where there would be multiple organizations that would derive value. Let's say, there's a pharmaceutical supply chain company that is shipping drugs around the country. So they may use your solution on their equipment. But then the equipment OEM, as you mentioned, might also derive value because that maybe indicates the maintenance issues that need to be addressed in the equipment. So they might see value in this. The hospitals that are receiving the medicines might want to have access to the real-time data so that they know when medicine arrives, that it's been transported properly. How does that tend to look where you're in situations where multiple companies might have value but then, of course, one of those companies is the customer, and the others are vendors or customers of your customer? Do you see a lot of data sharing between the borders of companies here?

Sammy: I'll answer, but I'll also let Gil talk about it. I think still, we are not — I'm not talking specifically SmartSense. I'm saying generally, we're still not at maturity in terms of data sharing between companies, specifically IoT data sharing. I think that there are some forces that would drive more sharing, like we see now on the Food Safety Modernization Act. They're pushing more for traceability. There's a Traceability Act that will be enacted a few years from now, where it will require companies to share that data. So that would then drive the sharing of IoT data to be more established, more streamlined between companies. But right now, it is still not. If you think about it, there's not really an infrastructure out there that will allow for that easy sharing of data, unless you're the big ones: the Walmarts, the Targets, and other companies, other retailers that build their own proprietary platforms.

Gil: Yeah, and it's not really a technology barrier at this point. It's more around the privacy of sharing the data from one client to another. There's just very strict rules around that. I think where we do see that coming together is more on the compliance side. If you look at the regulatory bodies as they traverse that supply chain, from the end product all the way back to where it was manufactured — exactly the value chain that you described, Erik — that's where I think this comes into play where they're going to start. Like Sammy said, there's a lot of movement in terms of understanding, okay, where did this product come from? Where was it stored? Who sent it to you? What were the conditions during the route or where it was stored?

I think from that angle, there's definitely a lot of benefit. Because those edge companies, if they're using something like SmartSense, they'll be able to provide that traceability even though it wouldn't be at the company level. It would be more at the product level. I think that's a start. It's still data sharing. It's still going through that value change, but it's more through a brand or a product rather than necessarily doing the entire product line for that supply chain, if that makes sense.

Sammy: Yeah, and I think every company in that ecosystem of IoT and the data that we collect, I personally believe that every company is obligated to have at least a portion of their solution open as an open platform. We've done that in the last year or so. We've announced that we worked on opening our platform to be able to get data out and put data in and operate in an ecosystem. If you're not operating nicely within all the other investments that companies have made, it's going to be very hard to survive. Companies are talking about data sharing. So you have to have your platform be able to serve not just the use cases that we were talking about — the users that work on the file — but other data to other areas, other insights to other areas within that ecosystem of technology.

Erik: Yeah, it's a really interesting part of the IoT. Because if you think about the consumer internet, it's not exactly open. But it's relatively open, right? You log on to this device, and you have access to just a seemingly infinite amount of information from different websites, and so forth. But the IoT is basically a bunch of closed systems that can be integrated with each other with great effort. Hypothetically, there's a tremendous amount of value if you're able to make seamless connections.

There was a company I had on in December called Narrative that's basically building a data marketplace. And so it's like you have different datasets. You track the metadata around them, and you can allow partners, supply chain partners, or customers, or completely third-party companies to either have free access to that data or purchase access with maybe certain rules about what metadata. For privacy reasons, they can't have the location, or they can't have certain things that are related to a specific location or individual. But they can acquire some part of that data. I think the interesting part of IoT. But I agree with you that this is still quite early stage, and it's not so much the technical challenge. It's the privacy, the regulatory challenges, and also the business model challenges. Businesses like to hold on to their data.

There's another aspect here, which for healthcare, I imagine is fairly important, which is this aspect of cybersecurity. Certainly, hospitals and so forth have been identified as hacking targets from people. Is that particularly relevant to you? I guess you have access to probably critical systems. I guess you're collecting data from them, but you're not operating them. So is cybersecurity a significant issue for you?

Gil: It's definitely something that we pay a lot of attention. One of the reasons why we have our model where we're not connected to the IT infrastructure is from a security perspective. Because let's say that, for some reason, the system is compromised. They're not going to get any access to your IT infrastructure, which is really the key vulnerability within IoT. If you think about industrial IoT in general, they're going to inherit all the same vulnerabilities that a typical cloud enterprise solution will have. But it also has one of the, I'd say, worst vulnerabilities which is physical access.

Like you and Sammy said earlier, the sensors are right there where the consumers are. A hacker could just walk in and pick up a sensor, take it home, take it apart, and try to figure out its protocol to the cloud and hack that. And so one angle is, IoT companies have to pay a lot of attention to cybersecurity and make sure they follow some really strict rules around protecting that infrastructure. But if you can separate that from your IT infrastructure, at least you gain that extra protection that if somebody stole that sensor and somehow made it in, they're still not in your network. They may be able to get some telemetry data, but they're not really going to be a threat. So yes, it's absolutely something that's applicable not just to us but to any IoT vendor in my mind.

Erik: Yeah, got it. That separation makes sense. Well, there might be a couple other points that you want to touch on. But one last point that I'd like to cover is this discussion of what's exciting about the future for you. So it sounds like you're investing in building this business. IoT keeps marching ahead with AI. There's also a lot of new developments recently that will also be impacting IoT solutions. But for SmartSense, what's around the corner? What are you most excited about in the coming year or two?

Sammy: I think you called it, and I would be remiss to not mention AI and machine learning and the stuff that we can do with massive amounts of data. Something that was a barrier several years ago is now possible. We're talking about billions of data points, and how do you sift through these data points to gather the best insights that you can have? We're heavily investing in machine learning. We're heavily investing in gathering insights from the data.

I think the data is really what makes us unique, right? You have companies. There are other companies that can provide sensors and can provide monitoring. But because we have access to such a scaled environment with thousands of customers and billions of data points, we can start providing more insights and benchmark that would help our customers. We get into analyzing the behavior of the equipment and how it behaves, and predictive maintenance. So many kinds of insights that we can gather and we can produce that would help our customers, again, to minimize their losses, to maintain their compliance. So the AI piece is one area.

The other area I mentioned, the open platform, we absolutely want to make sure that our customers, if they can think of something they want to monitor, if we think about something they need to monitor in order to, again, like we said, reduce the labor and increase the compliance, we want to do it. We don't want to get into, let's now develop a new sensor and another sensor and another sensor if there's something that is readily available out there. We want to make sure that we open our platform. We are able to bring in any third-party sensors and hardware that we can connect to our system, and provide value to our customers really quickly. So open platform is definitely another area of investment that we do.

Then finally, it's how deep can we get into the supply chain, into the monitoring the product, making sure that we can trace product. We spoke about reducing the costs, removing the barriers, to enter, being able to monitor shipments, boxes, small boxes. So what would be the cost of that? What would be the scale of that if you're looking at millions and billions of shipments every year? So really, it's about more insight, opening our platform, and creating that vast visibility across the build, the end-to-end visibility.

Gil: Yeah, and I'll just add because Sammy took all of the products once. Just from a technology perspective, I'm excited to see the direction that the IoT market is going in general. I think there's a lot of new protocols that are coming up that have better battery management and better connectivity. What that's going to do is help us drive the cost down, the form factor down. We'll be able to have more sensors and more places to do more things that generate more data that feeds into what Sammy was saying, in terms of the AI analytics and supply chain.

There's also some progress in different computing architectures which I'm excited about, like neuromorphic chips which will help us drive some of that AI processing directly to where the consumers are. So it's similar to how when you talk to Alexa, it's able to process your voice and react to commands. We want to do that at an industrial scale and make sure that we can drive those decisions as close to the consumer as possible.

Erik: Yeah, cool. That was my last podcast, which was just I think actually yesterday. It was with a company called Mythic out of, I think, Michigan. But they're doing analog computing. So basically, separate from digital computing but a structure that apparently is significantly more cost-effective for doing machine learning, and then with the goal of taking machine learning down to the edge. So I can imagine that as being something that Digi, in X number of years in the future, that you'll be doing a lot of machine learning of fairly sophisticated solutions at the edge for customers. I think really a fascinating set of use cases that are available there once we get to the point where we can run more complex algorithms on small compute. Cool. Guys, I think we've covered a good bit of the business here. Anything that we haven't touched on that's important for folks to know?

Sammy: I guess one thing that I would like to point out, and I've talked about it a lot. Again, this is part of the things that I've realized when I joined SmartSense and started working with our customers. We, as consumers, we have a certain level of trust in the companies, the businesses that we interact with — whether going to the supermarket to do our weekly shopping, going to a pharmacy to get prescription, medication, or going to the hospital to get some services. We have a certain level of trust that we put in these businesses. None of us are going to get into a restaurant and ask the waiter, "Hey, can you show me the chicken that I'm ordering has been kept in the right temperature within the restaurant?" Nobody goes into the pharmacy and says, "Can you show me that this bottle of vaccine has been kept at the right temperature?" There is a certain level of trust that we have as consumers in the businesses. The businesses are obligated to provide this safe and high-quality environment for consumers.

I think that as this, I would say, gap between the consumers and the businesses in terms of the knowledge of what's happening with this product, I think this gap is starting to slowly shrink. Consumers are more smarter today. Businesses want to be smarter. I really am excited to get to a point where it's not a talk about sensors. It's not necessarily a talk about data. It's a talk about that trust specifically, that I know I get into a place, and I know that it's going to be high quality and safe. It's similar to when we bought PCs several years ago, and it says 'Intel inside.' Everybody knew, oh, it's Intel Inside. It's going to be high quality. So that's kind of where I see the IoT getting into, more of the now, as a consumer, I trust the businesses because they are monitoring, and they are collecting information. They really want to serve me the best product that they can.

Erik: Yeah, interesting concept. So it'll be interesting to see, as companies start to invest more in this, how they actually start to brand and communicate that to their customers. It'll be an interesting development. So for the listeners, the website is smartsense.co. It looks like the best way is just, if anybody wants to follow up on this conversation, to reach out to the website. Any other channels that people should look at, or is the website typically the best place to start?

Sammy: Yeah, I would say the website is the best place to go. Definitely, follow us on LinkedIn. We generate a lot of content. We put up a lot of information on LinkedIn as well. This is where you can see where we are, what conferences we're going to, what are the latest trends in the industry. So definitely, LinkedIn is another channel for people to follow us.

Gil: Yeah, and we're happy to hear. Even if it's not just related to SmartSense directly, we're happy to hear anyone that has any other use case that they feel that this might be a good opportunity to connect. We always like to hear from our customers or potential customers.

Erik: Absolutely. You guys are a great platform for companies to collaborate with. Well, Sammy, Gil, thanks a lot for taking the time to talk to us today.

Sammy: Thank you, Erik.

Gil: Thank you very much, Erik.