- [Narrator] You are listening to the IoT for All media network. (upbeat electronic music) - Hello everyone. And welcome to another episode of the IoT for All podcast on the IoT for All media network. I'm your host, Ryan Chacon, one of the co-creators of IoT for All. Now, before we jump into this episode, please don't forget to subscribe on your favorite podcast platform or join our newsletter at iotforall.com/newsletter
to catch all the newest episodes as soon as they come out. So without further ado please enjoy this episode of the IoT for All podcast. Welcome, John, to the IoT for All show. How's your week going so far? - [John] Thanks for having me. Pretty good, how about you? - [Ryan] Pretty good over here as well.
I want to first start off and introduce my cohost, it's Shannon Lee. She's been on before. - Hello. - There you go. And John, do you want to start off by just introducing yourself a little bit, talk more about, you know, your background, your experience, kind of how you got into IoT, that sort of thing? - [John] Sure.
So just to introduce myself, so this is John Younes here. I'm one of the co-founders and COO of Litmus automation. So my main role is around go to market, sales, marketing, partnerships for the company.
And I'm based in our headquarters in San Jose, California. So I got involved in IoT about five and a half, six years ago now. I met my co-founder in a, maybe little bit longer actually. I met my co-founder in a master's program which was a global entrepreneurship program about seven years ago. And his background is actually coming from more industrial automation.
And obviously people have been talking saying we've been doing IoT in the industrial world for the last 20, 30 years with automation but it's a much newer phenomenon now with newer technologies like cloud and AI and machine learning and those kinds of buzzwords. So he was actually doing some more field level projects around connecting sensors on oil and gas pipeline, putting that data into a database and then figuring out, trying to predict when the pipeline was going to break. So that's where he had the idea, if companies had a scalable platform that could basically take data in from any sort of sensor and or device and make that data available to various applications that could leverage that data in a way they're looking for that could really eliminate the development cycle and allow companies to really adopt these kinds of newer IoT technologies and systems much more rapidly.
So that's kind of where the concept started from and where we got started from. Now we've obviously evolved a lot from that point and have two core products more on the edge computing side and enablement and management layers for IoT. - [Ryan] Great. Can you talk a little bit more about the opportunity you guys saw when you started the company? Moreso like how, what was the condition of the market? What were companies doing to, you know, I guess work around to solve certain problems you guys saw? And then obviously fitting in the idea to then build the company and get to the point of where you're at today, which is, you know, having these two products that benefit the industry.
- [John] Sure. So when we got started, we launched, I think in the market, our first version or first product about five years ago. And so at that point, IoT, it was a huge buzz word.
A lot of people started talking about it. A lot of people still didn't really know about it and what it actually meant. But it was this huge new thing that Gartner started talking about which is going to add $19 trillion in economic value and all these different buzz things around it. So it really was a huge new, hot topic but I think people didn't really know much about it or what they should be doing with it. So it was very, very early stages.
We've seen a huge evolution over the last five years of just kind of sophistication in customers and the market as well, where people actually now have an idea of what they're supposed to be doing. So it actually took probably about four years to get to this point where people have real projects, real requirements, where now they know how they're supposed to be leveraging these kinds of technologies. We started from the concept of more on the cloud side, which was just connect, manage, and integrate any IoT device. So more dealing with these newer IoT devices that were coming out, which there were hundreds of different types. And there was kind of no homogeneous way to manage these heterogeneous type of IoT devices and collect data and integrate it to more enterprise systems like CRM systems or databases or analytics tools. So we started from that point.
We started more general IoT, again, we were just trying out different industries, talking to different customers, seeing where, which kind of made the most sense for us to focus in. We had early success in the connected car space. So our first customer was actually Nissan and eventually working with Renault as well. So we did a series that actually took up probably the first two years of the business. That kept us quite busy working with those kinds of two big brand names. We found it challenging to actually expand in more of the connected car and transportation space.
So at that point, that's when we wanted to get more back to our industrial roots as a company. As I mentioned, my partner coming from Rockwell Automation and working more on the industrial automation side of things. And a lot of our core team also comes from that similar background. So we made a conscious decision about three years ago to really focus, get our focus back more into the industrial side. So it was a bit of a gamble at that point, especially when you have two key big names like these two car automotive OEMs to really move into a different direction at that point. But we did it for multiple reasons.
So it paid off greatly. It was a great decision. Obviously now, so about two and a half years, three years in the market, on more of the industrial side and we've built up now these two core products, which is this edge computing platform and industrial device connectivity, which is solving a huge challenge for companies that are trying to even get started with IoT. And then we focus more on the management of these edge computing devices now, how companies can actually, in a centralized way, manage all of their deployments and devices across multiple factories or multiple remote sites. So that's kind of how we got to where we are now.
And the, as I mentioned, I guess going back to your original question which was more around the market, it's night and day now. So we were working the early days with just POCs or people doing random R and D projects. And then they got into, finally they had some budget to work on pilots.
Now companies actually have large scale requirements and rollout plans for how they're going to deploy it across hundreds of factories or 50 factories. Whereas before it was just very, very isolated small projects that did not really have much fuel behind them. - [Ryan] Gotcha, that's great. Can you shed a little bit more light on the point you made about making that transition away from the auto industry into more of the industrial side? 'Cause I know like you mentioned, it's very hard to kind of leave big name customers kind of on the table and make that pivot. And I'm sure a lot of customers out there or companies out there might be in that same kind of spot where they have a couple of customers they really like, you know, big names bringing in revenue. But maybe they see the direction of the industry shifting a little bit for basically their expertise and their benefit, but they might be scared to kind of pivot away from that.
What was your guys' decision making process behind making that switch and kind of leaving those customers to be? - [John] Sure. So one of that is definitely product market fit. We had a product but the market wasn't quite there and it wasn't quite there for a few different reasons. But I think some of the main reasons behind that are the fact that a lot of, first of all, you only have a certain amount of automotive OEMs that are out there, so for a startup to limit the size of your market to only a select few, and also they're located in specific geographies.
So it makes it very difficult for a startup to tackle that. Also the nature of that market. So in the automotive market they have very long production cycles.
So to actually put something into production could be even four years, and for a startup, that's an eternity. Unless you're going to raise a lot of money or take a lot of services revenue along the way, it's a very difficult arena for startups to be successful in. So for those reasons, we wanted to go after a market which was a lot more wide open and can get value and ROI more immediately from the products that we were providing rather than wait sort of four years and try and navigate only a select few customers that are at various different stages or trying to just do it themselves. So that's why we couldn't really sit around and wait in that market and decided to pivot more to the industrial area and started really adapting and adopting our products for those markets.
And then now it took us probably about a year and a half or two years but we really started to reach that product market fit. And now it's very cookie cutter where we're able to launch into new customers at a much more quicker place where they can, even after even two meetings they're purchasing something from us rather than very long cycles to just get a simple proof of concept going and then having to collect more services revenues for four years, they can actually launch something into a larger scale production within three to four months. - [Ryan] That's great. Are you guys seeing more, not just engagement from customers, which I assume you are, but kind of from that first discussion you said you're seeing people buy stuff really quickly. What do you think is contributing to that change and that kind of, you know, I guess a couple of years ago people were definitely scared of technology like IoT, installing it with (indistinct) systems and disrupting processes for an ROI that may or may not be realized. So what have you guys seen kind of changed in your engagement with customers in the last couple of years to lead to them now being motivated or ready to purchase and install these solutions? - [John] So I guess going back to product market fit is one of the main points there.
We have something that they're really looking for and the fact that also we're not coming in from day one trying to pitch them the world. The fact that we can come in with a product which allows them to get started small and scale so they don't have to have a huge investment and put huge budgets and ROI behind it, that gives us a huge advantage. Basically, you can purchase a gateway in our software and off the shelf you can start getting started with your Industry 4.0 initiative in a matter of of hours. So the fact that we can get them to that quick adoption rather than having to pitch a huge project that's going to require multiple levels of approval, they can get started for a few thousand dollars instead of the half a million dollar project. So that gives us a huge advantage from that approach.
And then they can scale as they start seeing value. And then also it's the nature of the market as well. Now companies, they do have dedicated budgets to work on these kinds of projects.
So you don't have to wait 'til the following year where they have to put it on the budget and their budget approval and get budget approval for it for the next year. So that the fact that they already have these budgets and that they need to deploy them and work on these projects, that makes it move a lot quicker as well. - [Shannon] You mentioned Industry 4.0, can you touch on that and explain what it is and tell us, is it real? Is it just a buzzword? And can you some examples of how Industry 4.0 is impacting businesses right now?
- [John] Sure. So, (chuckles) it is obviously a buzzword but it's more about the next generation of, the next revolution in more of the industrial world. So obviously the last one came when there was more automation in factories. Now it's more involving people, process, things, and data. So the fact that, and trying to automate all of that in a much more efficient manner.
So this comes down to having better visibility over all your assets and people or things, better predictability over those things and kind of bringing everything together. So that could be anything from just having real-time monitoring of all your assets or it could be involving more types of new-age devices like AR and VR, which are actually coming to the manufacturing floor. Companies are adopting those kinds of processes. Trying to eliminate more manual entry of data on clipboards or just entry into Excel, now they have iPads where they're able to log things a lot more quicker and efficiently.
So yes, they are coming to the factory floor today. It's just more at what scale? Some of these technologies, as part of this, are being adopted much quicker. Some are being adopted much slower. But different companies are playing around with different parts of these technologies. So that's how I would sort of define Industry 4.0 but it kind of depends at what stage these companies are at their journey or which technologies make the most sense for them to adapt a little quicker.
- [Ryan] So when you guys are engaging with customers, the ones that you usually look like, are you, I know you have the two products you've mentioned. Are you working with them from a very early ideation stage through building out the full solution, like as a systems integrator as well? Or are you guys usually just kind of providing the different components through your products and then the other pieces of the solution would come through partners or other ways of engagement? - [John] So we like to be involved with customers at the very early stages because we obviously understand our product and the market and the use cases. So for us, it makes sense for us to be involved, especially at those initial phases. These first pilots and proof of concepts are extremely crucial because obviously if those aren't successful you're not moving anywhere from that point. So for us, we like, we do some integration implementation assistance services for the customer in those first pilot phases.
But once they start scaling this, generally we don't want to become a services shop. We want to remain a product company. So at those phases, that's when we involve some of our integration partners that can really help start scaling the projects in that factory or in global factories around the world who have much more resources and capabilities in that respect. So we like to handle that in the initial phases, but as the scaling up comes, that's when we involve more partners that are better suited for that and have more resources than us.
- [Ryan] And in those engagements has there been kind of a common issue or multiple issues that you've seen arise that maybe there's some advice out there for people who are looking to build solutions in the IoT space to better plan or think around before they come to a company looking to deploy and build a solution? - [John] Yeah, so I think one of the main aspects to that would be the fact that they should have, they should be solving a real problem or challenge. The problem, it should be driven, the solution they're trying to create should be driven from an actual problem instead of just trying a technology for the sake of trying the technology. So that's a pretty obvious one, but obviously, you'd be surprised how many people are just kind of doing random R and D projects for the sake of trying projects, the products, just because they have mandates to do so.
So they should first of all be finding, talking to the actual operational people and figuring out what what problems they're really facing and then find the right technology to actually solve that. I think that's kind of the first step. And then understanding what's the ecosystem of different aspects and solutions and pieces that they actually need to bring together a full solution.
So a lot of times companies, and obviously they'll talk to one type of supplier and they'll say, yeah, you're fine, just work with us and we'll be able to do everything but a lot of the times that's not the case. So really understanding the landscape is really important. And what are the technological pieces that they need to actually fill to get to that end solution is very important as well.
- [Shannon] What are some of the top use cases you're seeing in IIoT? - [John] So the first one comes down to remote monitoring or asset monitoring. So getting better visibility into your assets. It comes down to having, first of all, it starts from data collection from all your assets.
So you have to be able to, first of all, connect and collect data from all your different machines and then get that data to the cloud where then you can have a centralized dashboard where you can actually visualize how all of these machines are performing across all your factories. A lot of the times what companies are targeting from that respect is OEE. OEE is a manufacturing KPI which measures overall equipment effectiveness.
So they can actually visualize this across all of their factories and see how they're all performing and drill down at individual factories and see all the assets and how they're performing, monitor things like uptime, downtime, efficiency, failure rates, those of kind aspects, throughput. So visibility over the production is the first point. And then breaking that down into individual assets and machinery and then people move to more predictive maintenance. That's kind of the Holy grail that companies want to get to but you, first of all, you need a lot of data to get to that point. So once you have this data pipeline now to the cloud you need sort of three to six months worth of data where you're actually able to start seeing trends and patterns within the data.
And you have enough data for these machine learning systems that they can actually start to correlate and create and understand patterns out of the data. So then once they've done that and their internal data scientist teams have analyzed that, they create models around that data and actually run those at the edge. So now it's better to run these algorithms at the edge because you're taking quicker decisions, obviously, instead of sending everything up to the cloud and making decisions there. And also if you're dealing with control back to the assets that's always better to be done at the edge as well. And nobody wants to expose their factory to the cloud and have people that are able to push decisions back to the actual production and machinery.
So predictive maintenance algorithms should always be run at the edge. And that's sort of that next use case where it's when the data is coming in, either sends an alert to a local plant or maintenance person to come service the machine if it's likely to break down or if it's hitting some sort of critical threshold where it could break down any second, write back to the asset, turn off the machine and then notify somebody or even right back to the machine and auto-correct some specific variable which can lower the temperature of the machine and allow it to continue on what it's doing. So having that kind of closed loop control.
So those are kind of the key, I'd say the two most often used use cases that people are doing. Everybody wants better visibility. Everybody wants predictive maintenance. But there's many other use cases in terms of more looking at quality or traceability of assets through the process as well. - [Ryan] One of our recent guests was talking about ROI and how ROI differs in different regions of the world.
And I'd be curious from your engagement with customers in more of a focused space, what kind of ROI is like, leading the conversations that you're having? Does it always come down to money for most of these organizations and companies, or there are other ROIs that are driving their decision-making process on what they're looking to get out of their IoT solutions once they're deployed? - [John] So definitely ROI does play a factor into this. A lot of times it comes down to if this machine breaks, if it's a critical asset, if it stops my production run and is down for three, four hours on average it's pretty clear I'm going to lose X amount of dollars from not my production not running for three, four hours. So sometimes that can actually be hundreds and hundreds of thousands of dollars that companies are losing because they're not producing during those periods of downtime. So it becomes pretty clear at that point if you're looking at a critical asset. The other points that companies, not just ROI, but it's also the fact that if you're not adopting, everybody knows they need to be adopting these technologies.
And if you're not adopting these technologies as quick as or faster than your competitors then they're going to be gaining a competitive advantage over over you. And that's huge in manufacturing because if you can, the more you can lower your production costs the more you can invest in other areas in your business obviously. So that's super important for not just looking at ROI, but that's another big factor which is driving a lot of these companies is really the fear of missing out or the fear of being slower than your competitors who are moving at a much faster pace. - [Shannon] Is that why manufacturing is one of the most prevalent industries, ripe for implementing IIoT solutions or are there other reasons behind that? - [John] I would say so. It's obviously a hyper competitive industry where cost is a huge driving factor.
And if you can save even a few percent in costs that could translate into tens of millions of dollars in savings. So the ROI is a lot more clear in manufacturing and cost is always a huge driving factor within those businesses. - [Ryan] I have a quick question. It's sort of related but we haven't really talked about this much. But regarding the connectivity side of an IoT solution for industrial IoT projects, when you guys are looking at connectivity, what's the decision making process like if you guys are involved in that stuff at all? And then just looking at kind of a lot of the discussions around the industry now regarding LPWAN and the disruption that it has in different solutions, especially in IoT, I'd be curious to get your thoughts on what you think of LPWAN technologies and how familiar you guys are in using them in the solutions that you're advising these companies to deploy and build.
- [John] So those aren't so prevalent in the manufacturing world, or being discussed because everything is wired ethernet mostly, or serial. But now everybody's moving to more ethernet networks. So generally you're just using the factory ethernet. All these devices have been wired and connected for 20 years if not a bit longer. So those kinds of technologies are a bit less prevalent I'd say. I have actually seen situations, though, in areas that have manufacturing, like in Africa, where some of their machines are not even on the same network.
And then so at that point, you're looking at, okay, should I actually invest in actually wiring all of my machines? Which can be pretty costly in setting up the infrastructure and take a lot of time. Or I can use LPWAN and send everything over some sort of wireless signal to a centralized gateway and then get the data from there, which can be easier to implement and some more cost savings. So yeah. But most of the time you're working with those kinds of manufacturing, ethernet is mainly the communication that's being used. - [Ryan] Okay, that makes total sense. I guess just as we kind of wrap up here, we have a couple of last questions, one of them is just to kind of get your thoughts on the future of industrial automation.
And I guess relating that more to edge computing, kind of, you know it's seemed like more things are obviously shifting to the edge for so many different reasons. But just to get your thoughts on what is the future of industrial automation looking like as it relates to edge computing? And we just kind of see that shift even more day-to-day. - [John] Yeah, so I think it comes down to more intelligence at the edge.
So edge has been more about a means of getting data for now. But the next phase is going to be more about driving further intelligence whether that's better automating process or providing more value add on top of that data that companies are actually collecting and doing this all at the edge. Or actually having those kinds of closed loop control loops that I mentioned. So people are not really doing too much in the control aspect where they're there, based on the data they're writing back to the assets and actually taking decisions on that, or updating specific variables that will auto-correct the situation.
So I think that's the next phase that people are actually going to be using more closed loop control. There's a lot of apprehension and risks around that when it comes to things like security. So companies are a bit slow to want to do that.
Some companies are adopting closed loop control within these kinds of solutions. But it's still, I think, a very small percentage. So I would say that's the next phase.
It'd be more about driving this intelligence and more real-time control loops. - [Ryan] Very cool. So what can we expect to see from your company over the next six, 12 months? Anything exciting on the horizon that we should look out for? - [John] It's quite an exciting time for us. The company is growing quite rapidly.
And so we should be close to about 50 people by the end of this year, which is exciting. As well as from the product side, moving much more towards this latest trend that I mentioned which is driving much more intelligence at the edge rather than just being a means to get data. So we are adding more analytics features, more machine learning features within our product and all focused on the edge itself. - [Ryan] That's awesome. Well, that's always nice when you can get more resources to do what you guys envision, so I'm sure that'll be great, you know, wildly successful.
So congrats again. If there are, if our audience out there-- - Thank you very much. - [Ryan] Oh, absolutely. If our audience out there is listening and has any followup questions, wants to learn more about what you guys are doing, to pay attention to maybe just news stories, what have you, what's the best way they can connect? - [John] Sure. So you can always follow the company itself, Litmus Automation on Twitter, LinkedIn, we're quite active, @laautomation.
Or myself, you can find me on Twitter as well at @jyounes2 on Twitter or on LinkedIn as well. So we're quite active on social media. So definitely you can find us there or on our website Litmusautomation.com for further information and following our activity and press releases there as well.
- [Ryan] Sounds good. Well, we really appreciate you being on today. I think a lot of the information is great. We haven't had too many guests talking about IIoT lately so it's good to kind of get your thoughts and insights into not just your own processes and the stuff you're seeing, but also just, you know, overall what's going on in that space.
So we really appreciate your time. We'd love to have you back at some point in the future. So, you know, just keep us updated on if any new breaking things are happening with your company and we'd love to have you back on.
- [John] Great, thanks a lot. Appreciate it. And look forward to hopefully joining again in the future. - [Ryan] All right everyone, thanks again for joining us this week on the IoT for All podcast.
I hope you enjoyed this episode. And if you did, please leave us a rating or review and be sure to subscribe to our podcast on whichever platform you're listening to us on. Also, if you have a guest you'd like to see on the show please drop us a note at ryan@iotforall.com and we'll do everything we can to get them as a future guest. Other than that, thanks again for listening and we'll see you next time. (upbeat electronic music)
2021-04-14