What are the trends in the IoT enterprise platform market over the past 5 years? How has Vinay leveraged his patents in connectivity technology into a platform company that has attracted 4M USD of investment? Learn about the various market structures and situations in Asia and the US, and how this affects the development platform market. Vinay Nathan is the CEO of Altizon.

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 joined today by Vinay Nathan. Vinay is CEO and cofounder of Altizon, which is based out of the Bay Area, in Pune, India. Altizon is a company that focuses on the industrial Internet, and specifically on sensor data applications and SDKs. Vinay, thanks so much for joining us and discussing your market and your technology with us today.

Vinay: Thank you so much, Erik. Thank you for having me here. It's a pleasure to have this conversation.

Erik: Vinay, you're a fellow startup Leadership Program and then you have a background and quite a few other technology companies. Before we dive into all design and the market, share a little bit about your background. What led you to co found this company,

Vinay: I've been dabbling in this area for almost 17-18 years now. So right from my grad school days in LA in computer science, where I used to do research work on adult sensor networks, and then I had the opportunity later on in life to be a product manager for an RFID middleware product company. So in that sense, the enterprise arrival of what we term as internet of things today is has been a recurring theme, both on the engineering side of my career as well as later on the sales and business development management side. So it's been an area that has always interested me.

And I have spare particular interest in short range wireless networks, have a few patents on USB technology. And enterprises typically always need platforms; so we set out essentially saying, hey, let's build out an enterprise platform company that is focused on industrial IoT. We were starting off from India, so that is a huge manufacturing base over there. So that helped us get the product right, and the product validation going at scale under kind of rigorous circumstances and then that helped us as we expanded globally. I'm based out of the bay area these days, and we drive almost half our business globally out of our US office.

Erik: Back in 2013, I think there was still a lot of uncertainty around how the IoT enterprise platform market would evolve. I think, since then, we've seen quite a few companies enter, certainly startups, but also more mature companies building solutions for the space. How do you position yourself in this quite dynamic and competitive market? And how do you differentiate from other players?

Vinay: So as you rightly said, in 2013, what was considered to be an IoT platform play was quite different from what we have today. So when we started off, what was expected of a IoT platform was to do data ingestion and do mobile device management kind of functionality. So the early players really only had that. And we were among the first to bring about an analytics engine that coupled and expanded the definition of what an IoT platform did. So in the early days, it was about the fact that we were a big data platform from ground up dealing with unstructured data on one side and providing you with both in-stream analytics and the batch processing capabilities as a productized version of a platform.

And then today, it's a lot more than that. So there are three broad areas when you talk of enterprise platforms for IoT. One has to do with the edge for computing layer, everything to do with distributed intelligence and pushing intelligence to as close to the machinery that you're looking at connecting as possible. So that's one layer. The second layer is the core platform layer itself, which I mentioned earlier, which does data ingestion aggregation at scale and provides the analytic features to drive a lot of the machine learning and AI that people expect out of platforms today.

And the third layer, which is also very important, is the ready-to-go business app. So in our case, we build out a whole host of manufacturing intelligence products on top of a platform that provide a ready-to-go kind of solution for customers to work with. So all these three layers put together, there are very few players that have the depth and breadth to cover all of these three layers, along with the fact that they have built out an entire army of system integrators and channel partners globally that can help get machines and equipment connected pretty much in any part of the world.

So when we really talk of industrial IoT In particular, there are three main challenges that are expected of an enterprise IoT platform. First step is you need to be able to deal with the IT and OT integration. So you have centuries old machinery and infrastructure, in some cases, and you also have it, which is of several generations. How well can you merge that? Second is that you have to deal with the physical world. So you have a customer who may have a plant in four different parts of the world, how are you going to get all of that connected in a predictable manner, in a repeatable manner? And different players play at different levels of this. There are some like us that go end to end and provide the holistic outcome to the customer.

Erik: But before we dig into the technology, let's talk a bit about who you're working with? Where are usually engaging with your customers, is it typically that they have a well identified challenge or objective and then they'll evaluate potential platforms, they'll get in touch with you, and then you'll win a bid to an extent or are you engaging more in an evaluation of their operations and providing recommendations for where and how they can improve? How does it usually look in terms of the commencement of a project or a relationship with a client?

Vinay: When it starts off, typically, there's a conversation with the customer that the good thing is industry for auto and industrial IoT is already key goal that the CXO layer has around driving digital transformation within the company. So the conversation typically begins around what the priorities are for the digital transformation initiatives they have. And then in our case, we focus on the manufacturing activities, and prioritize those types of use cases before we go to some of the asset performance use cases and things like that.

So where we walk in with, as I said earlier, is the fact that because we have these ready to go business value apps that are out there, we tend to have a conversation either on the productivity module in that particular app on the quality module, the condition-based monitoring kind of asset based IP that we have. And what we have realized is there was a lot of dearth of ROI data in particular because you don't really know what outcome is going to be, because you have not done this in the past, or there isn't enough of a history of people doing it in the past.

So in about a year ago, we put out a smart manufacturing report that gives a very detailed view of ROI use cases and frames for implementation, trends across verticals like automobile chemicals and all of that. And that forms a great conversation area for us. And in most cases, while we're selling IP and subscription to our platform, there are some cases where we are in charge of the overall digital transformation initiative itself. And in those cases, of course, we provide a domain consulting to the customer.

Erik: Vinay, talk to me a little bit about India. So I'm based here in Shanghai. And if you look back at 2013, the discussion was probably very much around what is the internet of things? What is industry 4.0? And now it's matured quite a bit towards companies having practical implementation discussions. What does it look like in India if you compare the companies you're working with and the maturity of their executive decision making but also the maturity of the teams that are doing the deployments are actually using the systems that you're developing for them? What does it look like relative to your customer base in the US today?

Vinay: The dynamics of each of these markets are so different. For one of the key problems that are mentioned, in terms of dealing with legacy infrastructure, you may have a plant which is 100 years old, and there'll be maybe a British or German make machine that got built 20 years ago that are still working in this plant. And at the same time, you will also have the most modern equipment from one of the industrial majors sitting right there on the same line producing an output together.

We have data from almost 31 countries and machines across them that send data to a platform daily. So it's a very rich data set to analyze. And what's interesting, and going back to that experience is the fact that we had to really focus on this diversity of equipment upfront. We weren't at the privilege of saying, hey, man, I'll only deal with the most modern equipment and then I'll deal with this black hole equipment that sits in the corner separately because very often that may be the one that is driving productivity down.

So one of the things was early on in our life, we had to ensure that we had built something that made it very easy for you to get [inaudible 10:19] connected, that was a need for agents and intermediate hardware that allowed you to digitize signals in some cases and things like that. So we built a lot of IP around doing that. And that was kind of very specific to the India side of the equation. And the developing world at large, I would imagine has similar kind of situations.

In the North America market are the more developed kind of situations. It was more of a data silo problem. Yes, it is, of course, a diversity of equipment there as well. And very often our entry point was the fact that on our calling card, we said show us the most difficult equipment that you have and we'll figure out a way to digitize assets data that was important to the core app. So we did use that as an entry strategy.

For the larger problem, there's the data silo problem, and the fact that you don't really have a correlation happening between productivity data and quality data. So the functional requirements were a little different in that context.

The other thing that you mentioned is also very important in terms of the maturity level of the customer and what they're looking at. So, you know, growing out of India, it helped us build checklists and manuals that were very detailed in terms of being able to deal with staff that probably doesn't understand industrial IoT, but they have a control engineering background or something like that, that talks to their language and puts the data ingestion problem in particular, breaks it down into small pieces that they can handle.

So that was something that again, went a long way because in many ways, we took that same model to North America and there because at some level, it's 10-20 person more qualified workforce that you're dealing with on the shop floor as well, it just expedited the whole data ingestion layer of the platform and the adoption of the platform on the shop floor in a big way. So those were our learnings and experiences around dealing with these kinds of different market segments or markets and how one helped us expedite the other.

Erik: Before we move on to the technology, because you're working in such a diverse set of markets, what is the pricing or the business model look like? Because I imagine that when you're working with companies that have very different cost structures, you also have to be somewhat flexible in terms of your offering. So I know the answer here always is largely it depends on the project and so forth. But can you share a little bit around how the pricing model looks or how the business model would look like for one of your productized solutions?

Vinay: The core business model that we have is a subscription license that we provide for access to the platform and apps and it is tied to the number of assets that talk to the platform. So in the smart manufacturing case, it's a tiered structure based on cell line multiple plants and enterprise wide license. So, whether we talk of it in the context of India or Germany or the United States, the tiers are the same. The pricing model for the platform layer is also actually pretty much the same. The application layer, the business value apps, that is where we are in a position to look at some different sessions. Because what happens is the level of customization and support that may be expected in a market like India is different from what will be expected in North America. So we are able to provide differentiated service offerings and in some cases, also get it into an outcome based a model so that there is an upside to us against that. So the depends answer is really on the [inaudible 14:19] the rest of it follows a fairly standard process.

Erik: Now you outline briefly the different layers that you're working on the stack. On your website, I see these three different products that you're offering. So the tonus IoT platform to tonus manufacturing intelligence, and then to tonus edge, do those aligned with the layers that you identified earlier, or are you providing all three of those layers in each of these products, but each of these would be serving a particular set of use cases or customer profile?

Vinay: So those aligned with the layers that I mentioned earlier, so it's a one-to-one puzzle. The tonus edge is to do with everything, all assets that we provide IP assets that we provide that help you get machines connected and gateway products, intelligent gateways and things like that. IoT platform is the core layer in between, and manufacturing intelligence layer is the application layer on top of it.

So, when we talk of end users, there are two three kinds of customer situations. End users, what matters is the business value of being out there, you kind of sell against that value proposition and ROI that the customer can expect out of that. When you're talking of system integrators and OEM situation where we have connected product vendors that are out there that want to build a connected part of connected sensor network, for example, and a customer out of Utah build sensors for oil and gas, they wanted to build a connected sensor product.

So the core platform layer is typically an OEM kind of layer which helps us onboard customers like that. System integrators would want to build custom solutions on top of a platform. So they will also leverage the core platform layer itself directly. And the base layer, which is kind of the edge for computing layer is pretty much used by everybody. So there again, we have a dual model. Most of what we give in the context of the edge is typically available for free and downloadable through our self-service engine by anybody that uses a platform.

There are certain aspects that we monetize. For example, there's an appliance, which allows you to build an intelligent edge gateway, which allows you to, for example, to distribute a machine learning algorithm that you have perfected on the cloud version of platform and push it down on edge for local processing. That kind of complex application functionality is delivered right to the edge, then in that case, we'll monetize that. But that's the overall picture around the three layers and the stack as it is seen from a customer perspective.

It's kind of difficult to look into the future beyond the point. Traditionally, we've always looked at the edge as being the enabler. And so nothing prevents us from working with other edge offerings that are out there in the market. So the way I look at it data is an asset that is available for use depending on the context that a customer has or a partner requires and if there is a specialized offering that enables a particular market or something like that, we have no comps in terms of working with a lot of the edge appliance vendors and things like that. We look at them as natural allies to what we do. So the monetization is really cool to be driven more aggressively around the manufacturing intelligence suite and the core platform layer, then on the edge side of it.

Erik: Thanks for tuning in to another edition of the Industrial IoT Spotlight. Don't forget to follow us on Twitter at IoTONEHQ, and to check out our database of case studies on IoTone.com. If you have unique insight or a project deployment story to share, we'd love to feature you on a future edition. Write us at erik.walenza@IoTone.com.