Reduce Risk, Increase Productivity with Machine-As-A-Service

In this episode, we talk with Michael Cromheecke from SteamChain to discuss blockchain and machine-as-a-service. Michael breaks down how this model improves machine productivity and performance, with reduced risk and little up-front capital. 

“We use the same data management mechanism, the blockchain technology that enables cryptocurrencies to exist, but we apply it to machine-as-a-service. And what that creates for us is a record that can be shared between organizations, between corporations, between businesses in a way that’s transparent to all parties. That’s objective, the data that goes in, you know, it’s gonna be the same data three years, five years, 10 years from now. It’s resilient over time. And the big important thing is it doesn’t allow one party to restrict access from the other party. So both parties truly have shared ownership of that data. No one party can change it to the disadvantage of the other. No one party can turn off access to the disadvantage of the other.”

– Michael Cromheecke

Connect with SteamChain on LinkedIn


Announcer: Hi and welcome to Data in Depth, a podcast where we delve into advanced analytics, business intelligence and machine learning, and how they’re revolutionizing the manufacturing sector. Each episode we share new ideas and best practices to help you put your business data to work. From the shop floor to the back office, from optimizing supply chains to customer experience, the factory of the future runs on data.

Andrew Rieser: Welcome and thanks for joining us for season two of Data in Depth, the podcast exploring data and its role in the manufacturing industry. I’m your host, Andrew Rieser. Today we are joined by Michael Cromheecke, co-founder and CEO of Prior to SteamChain, Michael spent 17 years at Rockwell Automation filling various roles within the motions solutions divisions. Welcome, Michael.

Michael Cromheecke: Thank you, Andrew.

Andrew: Well it’s great to have you here. We’ve got a couple topics that we wanna dive deeper into, specifically around machine as a service and blockchain. Before we dive into that, how about telling us your story? Give us some background of your time at Rockwell and then what ultimately lead to the founding of SteamChain.

Michael: Yeah, absolutely, always a pleasure to talk about our business and talk about my background. But as you mentioned in your intro, I spent 17 years working for Rockwell Automation, one of the leading manufacturing technology companies in the world and had a lot of different jobs there. I started off out of college, got a mechanical engineering degree. And started off out of college supporting customers that were applying advanced manufacturing technology in their facilities. And it really gave me a perspective on the business of the struggles that OEMs and end-users of this type of equipment face when trying to achieve performance. And oftentimes it devolves into kind of a finger-pointing exercise where everybody’s looking to place the blame and defer the cost to the other party. And I always found it to be incredibly wasteful and not in the best service of anybody involved. I spent a lot of time jumping on airplanes and flying around the country and my role was a technical one so I was intended to help fix the technical problems, but often I found that those technical problems were difficult to solve because everybody was positioning themselves commercially on who owned the blame and the responsibility for the resolution of some of those problems. So later in my career I was involved running a business for Rockwell Automation in the area of advanced serval robotics and I saw another interesting phenomenon where new technology was always very difficult to introduce into the market because people were hesitant to take risks on advanced manufacturing technology without understanding the outcome that they would receive from that. So it’s a large upfront investment and a tremendous amount of risk on the performance side over the lifecycle of some of these assets. So kinda the two of those things, the early career experience and the later career experience, were both influential in leading me down this path of machine as a service, which I think is a much better alignment between the people that are selling this technology and the people that are trying to optimize its performance over its lifecycle.

Andrew: Yeah, fantastic, I appreciate the background and what ultimately got you starting SteamChain and I think you hit on a variety of topics there that we continue to see in this industry and some of the challenges they face so it’s always exciting to talk to folks that have been there, seen that firsthand and experience that but then are taking it a step further to create some of these solutions to solve for some of those hard problems that are coming out of the quote unquote industry 4.0 sector with all these digital transformations that are coming on. So let’s switch gears a little bit and just define some of this stuff. So I think most people are familiar with as a service model, but maybe you can just dive a little bit deeper into how you define machine as a service and how not only you but SteamChain approaches that and the value proposition back to the OEM.

Michael: Yeah, absolutely I’m happy to. And I’ll start with a really broad definition. We see as a service in general as being any business model where there’s some performance metric coupled with pre-agreed terms, contract terms that result in a financial transaction. In some of the classic business models that have successfully deployed historically. Everybody likes to talk about the office printer market where there was a time when people bought office printers and when they needed a new one, they bought a new one and when their printer wasn’t performing very well they’d have to eat the cost of that one and upgrade and get the new technology. But more recently and I think almost uniformly now, people don’t own their office printers. They sign a contract to a firm that manages those assets, ensures that they’re highly available and that they’re able to achieve the customer’s requirements. And the customer pays for that per page printed and it’s less of a headache for them, they don’t have to manage it, the company that they’re contracting with manages all that risk and ensures that they’ve got the right technology in the right place at the right time. And so that’s again measuring the performance metrics: availability, number of pages printed coupled with some pre-arranged contract term that may include the consumable products around the ink and the paper and whatever else is rolled in. And it results in an ongoing financial transaction on a monthly or weekly or quarterly basis. And so from a high level it’s a good example but these business models can be a lot more sophisticated than just counting sheets of paper. They can manage features on an asset, the ability to turn on and off capability, the ability to sell the asset as you normally would traditionally but charge some kind of rate for the post-sale service, based on the asset’s ability to achieve some desired outcome that the end user customer may have. And so when you get into manufacturing of course, the things that you wanna measure go up exponentially. The challenges and the variety of equipment expands exponentially, it’s not just an office printer that you can purchase from a number of manufacturers, it’s tens of thousands of different styles of machines that you can buy from hundreds of thousands of different suppliers. And so you need a very agile, very customizeable, very software-driven platform in order to keep track of that.

Andrew: Now I think that’s probably one of the better definitions and examples that I’ve heard so I think the audience will appreciate the breakdown of that. And so as it relates specifically to you all, can you maybe describe a little bit more about your all’s business model? So you kind of hit on OEMs, is the value proposition going to the OEM for you all to help them set up assets that they have on their shop floor or are you going to the OEMs that manufacture either machines, widgets, what have you and helping them create these as a service models for their end customer?

Michael: Yeah we work with OEMs directly to put together commercial programs so that they can begin to offer their machinery, their technology, to the marketplace based on new and creative business models that are a better value proposition for their end user customers, focused on performance, on outcome as opposed to the state of the market today which is machine manufacturers selling capital assets. Where really, when that asset is sold and delivered and the final payment is made early in the lifecycle of that asset, all the risk ends up in the hands of the end user. And all the capital ends up at that point with the OEM. And, you know, there’s pluses and minuses on both sides. For the OEM, that means every time they sell an asset, yeah, they get a lot of cash in hand. But ultimately they have to continuously sell new machines to grow their business and grow their market. It’s a very cyclical business and managing cashflow is a real challenge. For the end user, it means, “Hey, you know, we’ve acquired this asset “and we’re essentially taking all the risk “of the performance of this asset.” And other than the concern our OEM may have that we may not buy another machine from them. There’s very little that ties them to that OEM if they’re struggling with the performance of that asset. And so by looking at it from both sides and saying, “Hey, what if they entered into a relationship ‘where the OEM had a revenue stream “that was based on their end user customers ability “to achieve the goals, the metrics of this asset “over a longer time horizon?” It does two things, it transfers some of the risks of the OEM, but in the way that they’re getting paid to provide that post-sale service, a reason to go in and reason to monitor that asset, so that there’s more revenue and profit opportunity for them. There’s definitely incentive there. On the other side of the fence for the end user, it reduces the risk that they’re gonna be stuck with a machine that doesn’t achieve performance requirements. Or if they are, they’re gonna pay substantially less for that if a lot of it is based on performance. So we think it’s a better articulation that creates value for both sides of the equation because it truly aligns their interests where both companies when that asset achieves its optimal performance. And both companies face the same risks when it does not. And that’s the kind of relationship that I think everybody that is confident in their equipment, appreciates and wants to have because it’s true partnership at that point and we think that’s a win-win for the industry.

Andrew: Yeah, absolutely, couldn’t agree more. So this leads us to a good segue into a couple of additional topics. One in particular being the data and the metrics around what you’re describing. So obviously when you set up this framework and you set up this value chain and proposition around how the assets going to be measured and being a mutually beneficial kind of arrangement, then the next step in that is what are the KPIs or key performance indicators of how to measure the effectiveness of that for both sides, hence the data aspect. So maybe you can expand upon that piece a little bit and then we can segue that into blockchain.

Michael: Yeah, absolutely. So I think the complexity that you’re kind of describing there around what metrics to measure, that’s really, I think that the differentiator for industries that have already been impacted by as a service models. In manufacturing, it is significantly more complex because every machine may have different metrics. Every end user may think of value creation differently in terms of the performance, even of an individual machine type. You know, various customers may look at it different ways based on how they plan to use it. So one of the things that I think is really important to be able to do this in the manufacturing market, in the manufacturing setting is the ability to customize and configure and create multiple different business models for one particular OEM or one particular machine type that’s customized for each machine sold to each end user. And as soon as you get into that kind of complexity, the only practical way to do it from a bureaucratic perspective, the only way to administrate this is by using software. Otherwise the data just becomes too overwhelming. And so we’ve heard many stories of people attempting these things in the past, but generally they’ve done it without a platform, without an automated business process to manage it. Or they tried to build their own internal business process to manage it, and they get overwhelmed fairly quickly. And we think that’s the real differentiator with SteamChain is we set out to build a platform that allows you to deploy these business models and to customize and configure infinitely because there is so much complexity here, each one of these things, but in a way that is standard that allows you to audit, to review, to compare and analyze the performance of these contracts holistically over an entire portfolio. Because they all look the same on the platform and you can see exactly the details of the financial side of the equation, the performance side of the equation, and how the two things tie together with the specific contract that is managing a particular asset. And we think that’s a pretty powerful thing. And we think it’s the first time it’s really been attempted at scale. And, you know, we’re really excited to be working with a lot of different OEMs that are now beginning to I think they’ve always seen the value and the possibility here, but they are beginning to realize that now there’s a standard off-the-shelf way to manage the complex of this and they don’t have to invent it themselves. And they very much appreciate that.

Andrew: Yeah, absolutely. I love all these concepts and specifically around eliminating the excuses that were traditionally out there, right? So it’s breaking down all the hurdles and barriers that made this too overwhelming and too complex and abstracting that up in a logical fashion, whether it be from the upfront cost standpoint, the capital expenditure, minimizing that, creating the value on both sides, but then being able to round it all out and then tell the story with data so that those sides really are able to see the impacts of how this as a service is impacting the organization.

Michael: Absolutely.

Andrew: Cool, so the next topic that I’ll be brutally honest with you about became a buzzword a few years ago of blockchain and conceptually, I think I understand it, but it’s always been one of those things where it’s like, all right, what’s the real value behind it? And how is that driving services like what you’re describing? So maybe the 30,000 foot view definition of blockchain and then maybe tie it into the specifics of how this is being incorporated into platforms and services like SteamChain.

Michael: Yeah, this is always a really fun question for us because it has received so much hype and you know, everybody that opens a newspaper, has opened a newspaper in the last few years, you know, has read about blockchain, and a lot of that is pretty dramatic in terms of, you know, some people’s opinion on how it’s gonna change the world in so many different ways. And you know, I usually start any discussion about SteamChain and blockchain with a bit of a disclaimer. SteamChain is not a cryptocurrency company. We don’t buy, sell, trade Ethereum, coins or Bitcoin or any other cryptocurrency. And so that makes us a little bit unique in that we separate the concept of cryptocurrency from blockchain. And I think people that are kind of inside the industry understand what we’re saying there. What we do is we use the same data management mechanism, blockchain technology that enables cryptocurrencies to exist, but we apply it to machine as a service. And what that creates for us is a transparent record that can be shared between organizations, between corporations, between businesses in a way that’s transparent to all parties. That’s objective, the data that goes in, you know, it’s gonna be the same data three years, five years, 10 years from now. It’s resilient over time. And the big important thing is it doesn’t allow one party to restrict access from the other party. So both parties truly have shared ownership of that data. No one party can change it to the disadvantage of the other. No one party can turn off access to the disadvantage of the other. And so this mechanism is uniquely suited to manage the data between corporations for how they might share visibility and ownership of the data so that they’re both confident that what was collected on that machine and how it was converted into a financial transaction is exactly what they’d expect based on what they agreed to originally. And you don’t have to pay some subscription to continue to have access to that data. So it’s always available to you. You can confidently hook up your internal business systems to those databases and extract that data now and in the future. And it makes for just a very resilient solution for managing this data at scale. And as we looked at different ways to build this, we realized we didn’t need blockchain to build it, but building it with blockchain makes it so much more robust, so much more reliable and so much easier to administrator for us long term that it was just the obvious solution of a technology to deploy, to make these business models available in our industry. And so we moved forward with it, developed our expertise in blockchain technology and we’re very pleased we did because it continues to be just a great platform and technology to build on. And, you know, in the two years we’ve been doing this, we’re seeing more and more features, more and more capabilities that are just gonna continue to add to our ability to serve our customers in a way that they are comfortable and confident, that our platform manage to their data in a way that creates value for both parties.

Andrew: Nice, so circling back to a comment that you made earlier in the discussion, we’re hopping on planes flying into different locations and then really having to sift through all the noise and finger pointing. Does this help eliminate that as well by, because it builds in that insurance policy around the integrity of the data?

Michael: Yeah, you hit on it. Exactly, I mean, I would get off an airplane and you know, South Carolina and end up at a facility where the people that are operating the plant are saying, “Hey, this machine hasn’t run for three months “or it’s only ever achieved 20 units a minute.” Or whatever it is. And because that data was one-sided, there was no way to substantiate their claims. It was their perspective on what it was? The OEM would not believe it or say, Hey, the reasons why are this, that, and the other thing, there was no common data set that they all looked at and said, “Yep, we all agree that what was collected “is exactly right.” It was always influenced by, you know, those who collected it, the trust between the parties, et cetera, et cetera. If you end up in that same scenario and you’re all looking at the same data and you’re all comfortable and confident that it was collected in a timely manner, you know exactly what the machine did, when it did it, and you share ownership of that data. There’s a single version of the truth. Now you’re having a conversation based on objective data, collected real time that everybody had access to over each entire history and everybody will continue to have access to. And so the conversation has become a lot more straight forward. You’re not arguing about whose data is, right? You’re talking about how do we resolve the problem that’s very obvious in this data. And so we think that’s one of the elements that makes this a lot easier. And then of course, when you couple that kind of a shared version of the history of what happened, you couple that with the incentives that are created when you link that to a financial transaction, right? So that people are getting paid based on the achievement of those metrics, the performance of the machine itself, it really does realign the incentives and the interest in those companies to work together. They truly are partners now as opposed to currently there oftentimes adversaries. Which we think is, you know, in nobody’s best interest when you don’t trust your OEM and you’re not confident or working in your best interest.

Andrew: Exactly, so one last example that I’d like for you to share is something that you had mentioned earlier as well around just tying this together. So maybe you could walk us through the example of the smart contract and how the SteamChain platform works in that scenario between the supplier and OEM and just tell that quick story.

Michael: Yeah, so it’s pretty straight forward, right? So generally we work with OEMs, but we’ve seen this happen where it’s initiated from either side. So sometimes end users will say, “Hey, we’re looking to buy some machines and here’s, “you know, we want it to be linked to performance “and we’d like to put together a contract “and shop that with our OEM community of work, “or it’s us working with OEMs to present “their portfolio using these business models.” But it boils down to what’s the agreement? And often those things are based on metrics of performance. Frequently, they’re focused on throughput, right? The specification of a machine to accomplish in the packaging world, it may be cases packed. It may be, you know, bottles filled, cans filled, et cetera, et cetera. How many units per day, how many units per per minute, whatever those important KPIs are for the performance in the asset coupled with some measure of availability. There’s, you know, it’s not just did the machine run, it’s if the machine did not run, why didn’t it run? Is it because it was unavailable for technical issues or did the end user just choose not to run at that particular day? We often see quality as well as a metric. So how many good parts versus how many bad parts were accomplished on this machine. Sometimes we see very specific and unique metrics that are important to both parties. The amount of product extracted in a process step as a function of the amount of product input into that step can also be a very important metric in terms of the performance of the asset. So whatever is relevant to that particular machine, that particular application is what we would measure. And we would take that data. Let’s just use a cases produced and we would work on the financial model. So what is the payment on a per case basis? It could be as simple as every time it produces a case, it costs a penny. And we can measure that and manage that over time. It could be significantly more sophisticated than that. There could be tiers, there could be different steps, there could be different price points based on how the machine is used over time. There could be different price points whether or not these programs include the post-sale service of the machine, whether or not they include a warranty over the entire life cycle. So there’s lots of different ways we can construct agreements and we generally get involved on the front end consulting with and working with and doing the financial modeling of these agreements so that everybody has clarity in terms of what they’re agreeing to. And then we take that agreement whatever we come up with and we encode it into software, into our consortium blockchain solution. So that actually lives in a place that both parties can see that contract, not just the time they sign it and it gets enough, you know, sits in a file folder and when there’s a disagreement they have to go in and look and measure and calculate. It is actually the software itself that is the agreement and every piece of data that runs through that gets converted into a financial outcome. And they can see that bi-directionally they can say, “Here’s the payment I made.” And they can calculate it back through the algorithm to see what the performance is. Or they can see the performance and forward calculate through that contract through that algorithm and see what the financial results should have been. So they can constantly audit it if they choose to and they have access to that data over time. So it’s very transparent to both parties and as long as we work together to create a contract that is fair and equitable and agreed to by both parties, it’s a great way to to align your interest and make sure you’re getting what you’re paying for and make sure what you’re providing, you’re getting paid for as well from the OEM’s perspective.

Andrew: Sure, so this whole topic and conversation fascinates me. And one additional stream I wanna pull on a little bit is the financial modeling and the contracts. Are you seeing this as an opportunity for whether it’s the OEM side or the end customer side, to simplify and standardize around some of these financial models instead of making it overly complex or is the answer just really, it depends where, depends on the application, the asset, the KPIs?

Michael: Well, we’ve seen in the two years and what we expect to see as we continue to move forward is that the models are kind of converging, right? What we’re seeing are standard forms and we’ve created templates around these. So we have usage based financing template that allows you to provide machinery and get paid for it as it’s being used with no upfront cost and no transfer of title. We also see performance warranties where you sell the machine. But the performance warranty template is used to manage the performance post-sale providing, you know, spare parts and whatever may be involved in ensuring a high degree of uptime and availability and performance of that asset. We see things like feature management being very popular where instead of thinking about it holistically as a machine, the application of the machine as a service model to capabilities on that machine, the ability to turn on and off features and pay for those features based on how and when they’re being used. And we see it more and more managing the post-sale service aspect and creating a revenue stream based on not just, Hey, you know, today when a machine breaks down and the OEM provides service, they make margin on that. So their business does better when the machines break down more often. That’s not in anybody’s best interest. I think end users would prefer to pay their OEM to have the machine not break down. And so creating post-sale service models that focus not on a pain when you need service, but rather pain when you don’t need service are, you know, I think much better articulations of the value proposition that the end user customers would wanna see. And it creates opportunities for the best OEMs that are building the best, highest performance, most reliable machines to also be the most profitable. And I think that’s exactly what the industry would prefer. And so there’s lots of different ways to do this, but we’re seeing kind of standardization around certain models and we’re seeing that companies face the same challenges in deploying these things. And so once we solve them, it just adds value to every contract that comes next.

Andrew: So Michael, I really appreciate all the dialogue and conversation around this and more specifically about taking these concepts that I think can be overwhelming and can be buzzwords that are hard to understand. I think he did a fantastic job of shedding light on those and putting them into perspective of how this provides value within this industry. The last question that we typically like to ask is just any additional thoughts on this topic or where you feel the future of data and blockchain and machines as a service, and how it will evolve within this industry?

Michael: Yeah, that’s a great open-ended question. You know, I think one of the things that we’re really passionate about, we’re definitely passionate about industry 4.0 and the industrial internet of things, right? And I think there’s been a lot of companies that have been talking about all the promise that these technologies create when you begin to connect your machines to the network and you begin to extract this data, what they don’t talk about though is how you extract value from that. And other than in a very long range, you know, ideas like predictive analytics and machine optimization and so on and so forth. And we think all those things are absolutely spot on. But without some method to manage the transfer of value from one company to another without some transaction capability, we think it’s going to be very difficult for people to get full value out of their investments in IOT. So whether you’ve not yet made investments in IOT or you already have an extraordinarily sophisticated IOT strategy and have all your machinery connected up to the network, the next logical step is to say, how do we use that data to create agreements between ourselves and our customers, between ourselves and our suppliers, between ourselves and third party service firms that may be helping achieve performance on these assets, between ourselves and the people that may be providing the financial backing for these assets. How do we use this data to create value for everybody? Because we’re able to customize business models so that we can align the interest of everybody, so that everybody acts in the best interest of the performance of the asset. Because it’s not algorithms that are gonna optimize that machine. It’s not algorithms that are gonna extract the most value out of it. It’s the people in the organizations that have the knowhow and the capability. You just need to get them incentivized properly so that when things need to happen, everybody’s got skin in the game and everybody’s working together to achieve the results that I think everybody hopes for. And that’s where we see SteamChain being a real game changer.

Andrew: Fantastic, Michael, really appreciate your time today and thanks for joining the show.

Michael: Yeah, thank you Andrew, appreciate it.

Andrew: So for those of you listening, if you’d like to learn more about SteamChain and their solutions, I’d encourage you to visit That’s S-T-E-A-M-C-H-A-I-N.-I-O And if you’d like to connect with Michael, we’ll be sure to provide relevant links to his online profiles in the show notes. If you enjoyed this episode, please take a moment to rate the episode and subscribe to Data in Depth available on iTunes, Google, Spotify, Stitcher, and pretty much anywhere else you might listen to your podcast. Thanks again for joining us today. Data in Depth is produced by Mountain Point, a digital transformation consulting firm, focusing on the manufacturing sector. You can find show notes, additional episodes, and more by visiting Thanks for listening and be sure to subscribe wherever you get your podcasts.