Rockwell Automation, Inc. (ROK) Earnings Call Transcript & Summary

Hello and thank you for joining today's webinar. Before we get started, we have a few housekeeping items. The audio for this event will be streaming through your computer speakers so make sure your volume is turned up and your computer speakers are turned on. [Operator Instructions] Today's event will be recorded and we will be -- and will be available right after it is completed. [Operator Instructions] With that, I would like to introduce you to our speaker, Jared Comer. Hi Jared, how are you doing today?

Not bad, how are you?

Great. Thank you. Take it away. Floor is yours.

Thanks very much. That's a fantastic introduction. So hi, everybody. My name is, as you've been told Jared Comer. I am in information and analytics technical consultant here at Rockwell, and you can read a little bit more about me down in the bio. Today, we're going to be talking a little bit about one of the platforms that Rockwell has moved forward with, which is FactoryTalk DataMosaix. But more specifically, this is not going to be an introduction to DataMosaix, we actually have webinars for that, that you can go back and watch right. I encourage you to do so. This is more so talking about the concept of transforming industrial data into operational value. What we mean by that is fundamentally, how do we get value out of information using the DataMosaix platform? How can we make it valuable to our customers. And that's what we're going to talk about today. And so what we're going to do is we're going to be walking you through some of the steps that DataMosaix takes with a lot of your information. We're going to be taking a back-to-front approach showing you the end result and then showing you how we actually got there, and I'm going to be joined by my good friend, Chris Selph in a moment to show some of these demonstrations so that you can see it live in action, okay. Now what we're getting at here is how are we actually getting value out of this. Well, you need information, you need data to solve problems within your business, okay? That's a fact. There's no way getting around that. Today, you can either go and decide to replace a PLC or replace a drive and work on the risk factor. But that also -- that decision was made based on data as well. So where is it getting used? Well, it's getting used by reliability engineers who want to understand root causes. It's being used by process manufacturing engineers to understand the best yield or the best and most refined engineering process for a particular piece. It's being used by data scientists for predictive control and for prescriptive control of automation systems. But the biggest problem with a lot of this data is that it's messy and it's -- there's no way to really wrangle it. And so we get to the barriers of being able to utilize your information. You probably have in your business today, dozens of point solutions that say, oh, I'm useful for solving this particular pain point like maybe a CMMS, which is great. CMMS is a great way to start using -- utilizing your maintenance information, but that's a small piece of your business. You might be going after an OE solution that, again, only does a little small chunk of actually improving your business. You've got data in silos throughout your entire business, right? You've got all of your process information sitting in a historian somewhere that doesn't love to give it up. You've got all of your enterprise information like your recipes or your product information sitting in a SQL table in another area, they might be locked down by different security requirements. They might be running in different databases. It's everywhere. A big reason that customers can't really drive value from their information is because it lacks actual context, right? When you look at process information and you look at that tag and it says TT404, right, maintenance guy knows what that means. And maybe the floor guy does, but you move that any further back, I [ want to say ] TT404, starts to lose a lot of meaning. Starts to not make a lot of sense. That's why context is so important. Data not being trusted and difficulty scaling are 2 other big ones in this realm in terms of, hey, I'm not looking at what you're looking at. I'm not singing from the same tune. And the biggest issue that a lot of customers run into is, okay, I've done this for this small machine that I built. I've got that. Now I want to do it for this machine, okay, we have to repeat the process. These are the challenges that people are running into. And this is where FactoryTalk DataMosaix really comes in. It is the go between. It is the tool-set that you can utilize to unlock the value in your data. And I know that's a marketing term, but it's true. Fundamentally, you have all of this raw information sitting over here on the left. You throw it through the DataMosaix machine as if you were, you utilize the tool sets within this application to be able to add context, to be able to organize it, to be able to make it freely available and then bring that out the other side and provide you the easy and simplistic ability for scalable applications, visualization and so on. Now what I'm actually talking about, [ let's ] return over that. It's called DataOps. And what makes DataOps Industrial. Well, fundamentally, it comes down to scalable solutions that are easy to connect to that link the digital space with the physical space, right? We're talking about data operations where we have a governance model in place. Where we know where that information lives, we know where it's going. We know what we have and we know what we can do with it. And that's really a hard part for a lot of customers. They struggle with knowing what to do with that information. Every one of our customers turn us, go, we've got data for days. We've got DFD drives that give us 2,000 parameters. We've got OEM machines that give us tons of information, but they still don't understand how they're supposed to be utilizing it. And it's mainly because it hasn't been organized for them properly, hasn't been put in a way that they can actually utilize it. Okay? Now what we talk about here is we talk about scaling high-value applications, right? So the biggest thing the DataMosaix is going to help you do here, and that's what we're going to be talking about is really empowering you to scale a lot of these applications that are super, super valuable within your business, like enterprise reporting or wonderful historian capabilities and analysis capabilities on time series trending, asset monitoring, the gamut and I'm not about to sit here and read off of a slide, but there are a multitude of applications that need to be scaled to multiple parts of your business that would not be possible the way your data sits today. Now it might be possible, but it would be a heck of a lot of effort, enter something like DataMosaix that can basically provide you with the capability to do industrial data operations. Now IT departments and business management systems and financial systems have been doing this for years, right? Google has been doing it to optimize ads that are being read to somebody or understand analytics behind YouTube videos. So if the commercial grade and the consumer grade people have been doing it for years, why has the industrial system not jumped on board. Now is the time to push into that? And DataMosaix is a great way to help you with that strategy. You've got an edge management solution within Rockwell that allows you to get information, pull it from the edge, get it into a place that's usable to something like DataMosaix, which then is going to allow you to do some basic visualization charting, discovery, exploration and actual contextualization and organization of that data, and then you can move from there and make it available to all these massive industry solutions that Rockwell and others are coming out with. Okay. So like I said today, what we're going to do is we're going to walk you through those different steps in how DataMosaix actually helps you [indiscernible] these. And we're going to go from front -- our back-to-front approach. These are the 4 main steps of DataMosaix, okay? First, you make data available. You grab it got to come from somewhere, right? Full list of extraction tables that we're going to get to. Then you make data meaningful, okay? You start organizing that information in a usable way, and that moves you into the data useful category, making data useful allows you to start exploring that information, start seeing what ticks, what doesn't tick, what can be templatized, what needs to be changed. Then you move into the valuable piece. And valuable pieces into charting, industry solutions, analytics, predictive control, machine learning and AI solutioning, okay? So what we're going to do is we're going to take a back-to-front approach, and we're first going to talk about that last piece, okay? We're going to talk about how we make data valuable. And so -- what I'm going to do here today is I'm going to bring in my good friend, Chris, who's going to give you a bit of demonstration on making data valuable with FactoryTalk DataMosaix, talking about some of our third-party integrations as well as some of our charting capabilities. Chris, go ahead and take it away.

All right. Thank you, Jared. Hello, everyone. Chris Selph here, enterprise software solutions consultant out of Atlanta. Thank you so much for attending today. And I'm going to take you through a couple of things here around just how we sort of get started and how we make the system work well for us in terms of a use case. There's so much to talk about here, and we just don't have enough time to talk about and show you everything. And so if we -- if you're interested in this, we would love to have a conversation with you about your specific use cases and outcomes you'd like to achieve. But so first, I'm just going to talk about this particular use case and we'll show you what that looks like in the product itself. So this is a packaging line checkweigher use case here. And so we've got a couple of different types of data here, 2 different categories. One is real-time data coming from a checkweigher as packages come across that. And so you could to sort of make the leap to lots of different kinds of processes that you might have. This happens to be in food and beverage. So we've got some data here that is in our historian, and we have some production IDs. And we want to be able to give this data some context. And so we want to be able to -- to give that context as well as answer a lot of questions that people are trying to -- working hard, trying to sort of build scalable applications across the enterprise, and it's very difficult to do. I know I've been a part of that for many years. And so you want to be able to answer questions like what was the total weight, the salable weight, the rejects, et cetera. And so we want to be able to see how -- see what that looks like. And so I'm going to take you to the sort of what the solution looks like here, and this is where we use that extractor concept that Jared talked about a moment ago, we're going to see more of that later around the -- getting the data that we need to be able to answer a lot of these questions. And so that comes from our production MES data as well as our historian data, and we bring that in and ingest that into the platform and then begin to apply some logic to that to do our calculations and store that information and even -- as you see down at the bottom here, we had some other enterprise applications that needed to consume this data as well. And so we're -- we can egress out to other platforms like MQTT and Kafka, et cetera. And so what does that look like sort of on the boots on the ground, and we're going to kind of start off by -- I'm going to share my screen now. And just bear with me for just a second here while I get that. And so you should be seeing my share now. And so this is one of the applications that we can really start taking a deep dive into our data. And already off the bat, we see some of the contextualization happening here right now. So the -- you can see the checkweigher data that you see here in blue as well as this light green. These light green boxes represent the production events. And so already, we're seeing information about our production here. And so let's talk about some of the other things that we can see here. So we know we have these events here. And so -- these events represent production -- the execution of production orders. We can see even more detail around those production orders here as well. So we can see what product was being made as well as a lot of other interesting aggregated metadata statistical information about the product as it came across the production line. And so if you think about the situation that you might be in where -- maybe you have a quality issue and you're in a containment situation. This is the way to get the -- get to this information very quickly so that you can execute the containment and make sure that bad product doesn't get out the door or maybe your compliance situation, maybe you're under an audit right now, and you need to have this information at your fingertips here. So this is a really a great way to do that. Now if we zoom in here a little bit and get a little bit more granular look at this data, so we can sort of have a look at this and we may be more of a process troubleshoot -- what's going on with our weights. And so perhaps we need to be able to see thresholds around this, so we can add thresholds so that we can see exactly what the data is doing and then also be able to do -- take it even further to do process monitoring where we set these thresholds. The system is going to be trapping for those excursions. There's process abnormalities and logging events associated with that as well as sending e-mails, et cetera, and being proactive and having the system do the work for you instead of forcing someone to come and look at the visualization and then make some decisions about that, okay? So very important things that you need to be able to do and have the system do those kinds of things for you. And so I call this -- a lot of times I call this more of a data exploration tool than a -- than just a simple visualization tool. And so here's a quick example here. So we can create new interesting data here. So what we'd like to do, there's a lot of jitter in this data here. So maybe we want to be able to do -- create an average associated with that. So we can do some calculations here. There's -- first of all, I'm just going to say, I want to -- the Checkweigher weight is going to be my source and then we're going to add a function. There's a very large library of functions here that are available to you from a mathematical perspective. And so perhaps we want to do something around -- doing a simple moving average, so we can just add this moving average in here. I'm just going to wire this up here, and then we'll take our output and wire up to here. And so that's going to calculate our average for us. And so then we can -- maybe we want to make that stand out a little bit more. So there's our moving average to give us a better idea about what's happening with that data. Notice here we can save and schedule this so we can actually create a new synthetic tag, if you will, here, which represents the average weight here so that can be used with other analytics, et cetera, that you would want to be able to do with it. Now so let's just think about things from a different perspective now and the different kinds of roles and users that you have in your organization, perhaps you need more of a dashboarding kind of view for the kind of role that you're in, you need the information to be aggregated up and displayed to you. And we have even more information here now. Remember, we just had the weight coming in, but now we have a lot of really good actionable information here such as the total line throughput, the salable product, the amount of waste that we've had, so if you're very interested in understanding whether you're giving the end-user the stated product specifications or you want to understand about giveaway or whatever the case may be, then you have this information now at your fingertips. So we've transformed that single weight that's coming in from that Checkweigher into a lot of actionable information. So down here, you see the production orders as they were executed, how much production was made during that particular time. And we can show here, we can see what's happening right now. So these are live values coming in from that Checkweigher, so you can see information around that as well as some historical information. Here are completed production runs down here. These are production runs that are currently in progress, et cetera. So you're getting a rich set of data here, and we can even drill down into this information and get even more granular detail around -- these are process upsets or abnormalities that we see here in the data here, process alarms and so forth. So a lot of rich information that's going to be made available here. So hopefully, this gives you an idea about the sort of the visualization and what you could do in terms of your production meetings, how you could change the complexion of those meetings and come to those meetings and are more prepared to be able to take action and solve problem. So with that, Jared, I'll send it back over to you.

Thanks a lot, Chris. That was fantastic. So one thing I want to point out to people is the visualization capabilities that you have within that tool set or just within mosaix. Now it was. There was some fantastic tool sets that are very visual, very no code within the DataMosaix solution. And it's not just an organization solution and a lot of people can get by just with the capabilities of the charting solutions within DataMosaix you look at being able to drag-and-drop calculation capabilities and overlay events. There are a lot of high-value IoT solutions that can hardly do that. But one step further, if you need even more, if you need more dashboarding style solutions or say you need to bring new information into something like an analytics notebook, something like Jupiter or Azure Synapse. It's an open API. That information is still there for you. So many people get terrified by the concept of, oh, my information is going up to the cloud. That's not necessarily going to be useful. I can't get back to it. Well, it's more open than anything. And the beauty of it is, when it's up there, well, you're not paying for the compute power to keep that rolling. That's a big thing, okay? And we're going to talk about that in just a little bit. So I want to reiterate, you have a full capability of doing ad hoc analysis for your domain experts and everybody else local to the system. You also have the ability to deploy those applications to more scalable apps -- IoT applications throughout your business. So for instance, if you've got other people coming in for instance Gartner, you've got PTC, you've got all these other solutions that your enterprise accounts may have standardized on. Power BI is another great example of this or Grafana you pull those in. You can bring that information in. It is still available, but it's better organized and easier to utilize. So we're going to move to the next step here. How do we make data useful? So we've moved and looked at, okay, how we can look at it, how can we visualize it. Its all well and good. You've told me, how great it looks. But I know a lot of different tool sets that can do that anyways. Let's get into the powerful piece -- of how do we actually make data usable? Meaning what does DataMosaix do to allow me to actually use this information correctly. We're talking about things along the lines of we're going to talk about PID drawings. We're going to talk about asset hierarchies and things like that, building out these data models within your actual application and data exploration. And with that, I'm going to let Chris take over and give us a little bit of an example of the data exploration and asset hierarchy capabilities within DataMosaix. Chris?

Yes. Thank you. Sharing now. Okay. Very good. So as Jared said, we're kind of working backwards. I showed you sort of the end result with the charting and the dashboarding. And so now we're going to get into the concepts around the actual data model itself and what enabled us to be able to do that data aggregation, that analysis, the visualizations, et cetera, and so let's just talk about the data model. And again, we will eventually get to the point of talking about how we populated that data model. But for now, we're going to have a look at the data model itself using this Data Explorer application here. And so you have the concept here in the platform of being able to create what we call data sets. And so this is -- it could be private applications that you're building, depending on the -- your particular role and the kind of problems that you're trying to solve and so you can create these data sets that sort of put some boundaries and [ guide ] rails around the data that you are that you're working with. What we're seeing now is sort of a global look at the data model itself. And so you can see the different sort of industrialized components of that things that are really focused on industry, which I think is a big differentiator for us with DataMosaix as we are a very industry and manufacturing focused here on our platform. And so you can see the concept here of assets of time series, events, files and so forth, all the different things that sort of make up an industrial application right here at your fingertips. And so we could -- we can have a look at our assets. So something that's really important is this concept of an asset hierarchy. And so you probably have lots of hierarchies that are built in various systems in your enterprise. But they probably don't match. You might have one in ERP. You might have one in your maintenance management system, one in your quality system. They probably don't all match and so this is a way to bring all that together in 1 place and have a hierarchy that everyone can then used to solve problems with. And so we can drill down in this hierarchy, and then we can have a look at the hierarchy, the different areas of the plant and then different assets in the plant as well. So when we look at this particular asset here, this happens to be a compressor. So we can see the constituents of -- that are associated and contextualize with that particular asset. So we can see time series data, we can see events here, we can see files and so forth. So we kind of get an idea about what data is really available to us. And even get some insights from this as well without using other tools. So depending on your role in the plant, you're going to be able to get insights from this as well without really relying on some other reporting tool or dashboarding tools. So this has value in and of itself. So we can see things like time series data, so we can see the data that's coming in. We see some thumbnails over here on the right that give us an indication that yes, we have data here around this. And so we can have a look at this data and then we can even -- we have a basic charting application here that we can use or we can open it in the proper charting application that I showed you a few minutes ago. And so when we look out and look at other things that are going to bring value to the application, we have this concept of files. And so here's -- one of the things that I think a lot of manufacturing facilities suffer from is that during the design process, there's this P&ID diagrams and other kinds of diagrams that represent how the actual plant was built. But oftentimes, those files, that data, that information is not really used. And so what we'd like to do is, be able to show this information in a way that brings context to the data that you currently have. And so we have used our AI under the hood here to now start matching the information, the data that you're seeing in terms of your assets and the instrumentation and sensors in your plants. And so for example, we can see here's our reactor and we can see a lot of information about that, including all the time series that are available to us and then drill down from there as well. And so we can also see that in action back over here on our chart. So if we want to add another tag to this particular chart, we can see we have the same P&ID concept here so we can drill down into -- excuse me, so we can drill down into this and add a particular tag that is of interest to us. We can use that to browse our available data based on the concept of the P&ID. Okay? And so those are the kinds of things. From a data modeling perspective, you want to model everything. You want to have everything available to you. And so those -- all these concepts are really critical to be able to not only provide you with visual components like you see here, but also to be the source of any kind of machine learning and analytics applications that you want to apply. So you -- we've sort of transformed that data from your source data. We've cleaned it. It is now a good clean source of data that you can apply to any sort of analytics applications as well. So it really kills 2 birds with 1 stone there. Okay. So I'm going to leave it there. And so Jared, I'll throw it back to you.

Awesome. Thank you very much. So we've now discussed about how to make information and our data usable through visualization. Now we're looking at how we make it useful by applying asset hierarchies and being able to organize it in a sensible manner. Right? So how did we get here? Now this is really, in my opinion, the biggest tool set of DataMosaix, which speaks towards making data meaningful. This is where you get the core capability of this tool set, which is allowing you to create what are called industrial knowledge graphs. Now what does that mean? Fundamentally, every piece of data within your business has connection to other pieces of data, okay? That probe, that temperature probe on that milling machine. Well, that milling machine, that temperate probe attaches to that milling machine, it has a relationship to it. That milling machine has a relationship to that line. That line has a relationship to your enterprise information because all of the information around that particular line could feed into, okay, how does that line producing, what's the production metrics, what is the recipe running on that there are relationships everywhere and trying to physically map out those relationships could take ages. And it's the one thing that prevents a lot of people from actually opening and fully taking advantage of their data because making those relationships for one, there's no tool set to do that. Normally, you have to hire somebody. You have to hire a data engineer or an architect of some sort to be able to develop those things. And even then you don't get the capability of having all of that contextualized data in one place afterwards. It can be extremely IT-heavy, and it can be extremely effort-heavy. So what if we had a tool set that could allow us to -- first of all, contextualize that data, right? Again, we have a milling machine that has multiple different sensors on it, but also has work orders that tie to it. You also have 3D model information that's tied to it. You've got maintenance information, hardware specifications. All of these things that surround a piece of machinery and data contextualization is so vitally important because just calling it MT606, well, that's not a unified name, right? That's not something that everybody across your business agrees, hey, that's the name of that. Right? MT404 technically represents the milling transfer unit in Plant 4. And when you start getting up to the enterprise level and you start contextualizing information of all your plants you start to see a lot of repetition and you also see a lot of confusion. So this where data contextualization is so vitally important. And it's one of the main reasons why DataMosaix is here to help with almost specifically that issue. And so what we're going to do next is I'm going to have Chris walk you through the truly powerful data contextualization tool sets like the automatic entity matching for DataMosaix, which in my opinion, is probably its biggest powerhouse. So go ahead, Chris, one more time.

Thank you, Jared. Now to start here guys around this concept of entity matching. And so what has transpired so far in our system. We have ingested data from a lot of different kinds of data sources. And so we have time series data, which could be tens, hundreds, thousands of time series data points that need to be contextualized with and contextualize with the assets and match to the assets that they're associated with. That's going to help your end users tremendously and be able to find the data and solve problems with that data. We also are ingesting other kinds of data like it could be quality events. It could be alarms and events from your process. It could be production events that we've already seen here today. And so the idea here is how do you create that contextualized data model. And one of the ways that we do that is using this concept of entity matching. And so this -- we're not saying that you should have to go and build this by hand using lots and lots of excel. Well, let's use the in-built AI algorithms that are built into this system to be able to do that for you. And so that's the concept that we're talking about here. And so we have this concept here. I noticed there's this concept of a pipeline. You're going to see that throughout the product. So pipelines are really just automated sequences of events that happen on your behalf on a scheduled basis, that is going to take care of these things that need to be taken care of or execute it on a periodic basis. You're going to see that with data ingest with the transformation concept as well as here on entity matching so that you can be sure that everything as data is being ingested, is being -- it is being matched as low as possible. I'm going to take you through a manual process here. And so let's just say that we have lots and lots of say, thousands, let's say, of time series data that's associated with this particular data set here. And so we want to be able to match these tags with the assets that they are currently associated with. And so -- the way that we do that is we kind of configure the system to do what we are -- we want to do here for the specific data that we are interested in. And so I'm going to select now the assets that the AI should be considering when doing the matching here. So we've selected our time series and now we've selected our assets as an example here. And so we can go here. And so the idea here is that as humans, we understand the existing data structures and the structures that we've created. We want to give the AI as many hints as possible here. And so we want to be able to say, look at this, we know that we've created this data, this -- perhaps even this metadata that's associated with this. So that is going to help the AI. And so for example, we can come here and choose some metadata that we've created around this data -- about the data. And so we can have a look at that and say, look at this and see if you find a match in the time series with the asset itself. And so now we're going to have the AI actually run the model here. And so remember that we're taking care of potentially tens of thousands of points here and without having to do any manual wrangling of data in CSV files and all the rest. And so what we're seeing here is the results of that. And so you can see different pattern matching that is happening here and then its confidence scores around its matching. So we see that some of these are 100% matches, which are great, and we have some that are lower and some that are even lower than that. And so if we sort of have a look at the one that's 100% here. We can see that it found in this name, this 904 matches with the asset named 904. And we say, yes, that is good, so we can confirm that and so for ones that are lower than that, so we see some that are -- that it's trying to match, but it just doesn't have enough information there. And so we know that your data sets are not going to be perfect coming in, but we do provide you the tools especially around transformation so that we can create the right metadata so that we can then feed this AI system to be able to actually make a good match here. So we have tools that you can use to mitigate this particular problem. And so once we've done this, then we can -- then we see, as I was showing you a few minutes ago, we can now see the assets, the -- and then the time series that should be associated with these assets, the events that we brought in from different systems that are associated with these assets as well, okay? So this is a really big timesaver in terms of being able to populate the data model with good solid data. Okay, Jared. I'm done here. I'll send it back to you.

[Audio Gap] Data contextualization. I hope everyone understood from what they just saw the inherent power behind that. You can feed tons of data into a solution like this into something like DataMosaix and it will then transform that into usable information by saying, hey, I'm putting my hand up here, this is essentially a data assistant for you. It's somebody putting their hand up and saying, i think this connects to that like this. And if you're telling me that, that connects to that like this, can I assume that everything else does. And if it does, hey, I've just done hours of work for you. Hours of work for you. There's a particular piece that I want to point out around the data contextualization piece because it's probably one of the bigger things. We're going to talk about that in just a second. But the next thing we have to talk about is, okay, that's all well and good. DataMosaix can do really good things of my information with my data. It can contextualize it and get it ready for usability. How do you get it in? I need to get there. Right? And so now we get and move on to the next piece, which is making data available. Now one thing we struggle on in the industrial space is a clearly unified way of pulling information out of our production area. Okay? A lot of times, we have to rely on proprietary solutions or lockdown of PC solutions. And those are good. The problem is nobody has really come out with a super flexible solution for getting that information out. That's where DataMosaix really comes in and it's extractive solutions, right? Being able to pull time series information, not just from Rockwell historians, we're talking [ wonderwork ] historians. We're talking a lot of other different third-party historian solutions, a lot of other third-party database solutions, engineering diagrams, 3D models from STLs to SVGs, you name it if it's information regarding your business, it can get in here quickly and easily. And so for the final demo, we're going to sit here and just show you a quick bit of information around the multitude of different connectors this has as well as what it might look like raw. So Chris, go ahead and do -- give us one more show.

You got it, Jared. So as Jared was saying, we have the concept around bringing data and landing data onboarding data, if you will, into the DataMosaix platform. And so that's a really important part of the data journey, if you will. And so we're kind of starting -- this is at the beginning, we saved the beginning for last year. And so when we look at our -- under our integrate section here on the platform, we have this concept here of connecting to source systems. And this is where you would go to find the exact, what we call extractor that would connect to your source system and then publish that data into the cloud, into the DataMosaix system. And so we have a lot of different extractors here. I'm certainly not going to go through all of these. And we -- what we see generally, especially in the beginning our extractors in essentially 2 different broad categories. One is SQL-based systems and others are -- other sort of broad category would be real-time streaming data such as OPC UA. And so we have -- we see here a database extractor. This is kind of a generic database extractor, but we also have extractors for other kinds of enterprise system, you see one here, for example, this is the SAP extractor. So in other words, we can integrate directly into the ERP, MES, et cetera, that might be running on the SAP platform there as an example. So you can see quite a few of the -- there -- it's really difficult for the data to hide from these DataMosaix extractors here. And so you see one that's probably very familiar to many of you. This has KEPServer here. So again, there's a lot of extractors here that it's probably going to be possible to find one that's going to fit your need. And these extractors -- what's -- they are actually running at the edge. These are edge applications. And so we can deploy these extractors in a lot of different ways and sort of help you manage those from an enterprise perspective. And so I want to make sure that you understood these that we have the data here -- sorry, we have the capability of accessing your data from wherever that data is and ingesting that data into the platform. And typically, that data is then -- that process, we talked about ingest pipelines a few minutes ago. So that pipeline says, okay, we're going to connect to the SQL Server database. We're going to execute a query. The results of that query are going to land in our staging area and then from that staging area, we may apply a transform -- a transformation to query that data and then maybe even create more an interesting metadata associated with that and land it in its home where it can then be visualized and analyzed. So that's kind of the first process here to kind of get that data model populated so that you can then use it. So with that, Jared, I'm going to toss it back to you

Awesome. Thank you, Chris. So once again, let's take that back to front. We've been able to visualize the information and utilize it in scalable applications. We've been able to make information meaningful by actually transforming it and contextualizing it, putting it in a space that actually makes sense for whoever is trying to look at it. We are able to contextualize and organize that data to from absolute raw information to actual usable information. And then finally, we're able to pull that information in. Now that whole step, that whole journey can take a long time if you don't have tool sets like this. Right? The thing is, is that you are able to -- with Rockwell specifically being involved with this DataMosaix solution, you're able to create -- we're able to create scalable domain expert initiatives. We're domain experts in the industrial space. There's no 2 ways about it, right? We're able to help simplify access to industrial information. We're able to provide industry-specific solutions. We work with mining. We work with food and beverage. We work with CPG. We work with all of these different industries. We have experts in every single one of them. I myself, I'm an integrator of 15 years. And I get to pride myself in being able to provide that kind of experience in the solutions that we create. And our solution partner ecosystem is absolutely massive. Right? Now one thing I want to end on and just I want to sort of mention here is a fantastic slide that talks about the time to first value of a solution like DataMosaix. When you look at all of the effort it takes to go through what DataMosaix can actually do for you, the time savings are clear, 200-plus hours potentially. And once you start scaling those out to multiple applications or multiple sites, well, that time savings becomes even higher. It's an effort upfront. There's no doubt about it. But it is a significantly more reduced effort and it gets customers on their way to digital transformation in their business. I want to point out digital transformation is not a thing you can do overnight. It is a fundamental reset of your mindset within your business. Making sure that you are being thought forward, data-centric and making sure that every bit of information that your factory or your production produces, it's going to broaden value back to you. And that's where DataMosaix can help. Now if you want to learn more, feel free to reach out to an adviser. Feel free to reach out to a Rockwell employee, and they can navigate you. This recording is going to be available for everyone as well. But Other than that, everybody, I just really appreciate everyone being here. And I guess, Raj, I think we'll move into the question-and-answer portion of this, Raj?

Yes. Thank you very much for this lovely presentation. It was really useful to see all of those features live. [Operator Instructions] So I will start by a few questions that came that just came in. [Operator Instructions] So first question is, this looks like Cognite. So how is this different than CDF?

That's a great question. CDF as many people might not know Cognite Data Fusion is the solution in which DataMosaix based off of. Consider this -- it is Cognite Data Fusion future, fundamentally, but it's got unique Rockwell capabilities. You can think of this almost like an OEM relationship where we are providing -- Rockwell is providing a lot of the capability that we come from a lot of the experience we have to improve upon this solution and really bring it to our collective customer base, right? We intend to build on the base of CDF in areas of configuration, deployment and management of the connections, especially for Rockwell data sources, as well as developing industry solutions that utilize the FactoryTalk DataMosaix platform. So fundamentally, it is CDS. But Rockwell solutions set DataMosaix is going to be really geared towards helping the industry gain advantage from this product.

Okay. Sounds very good. There is a question here that came in. Is Rockwell selling Grafana, Power BI, et cetera?

Yes. So Rockwell, I mean, obviously, we are not about to turn around and become an OEM vendor for something like, say, Power BI, Microsoft doesn't love us that much sadly. And Grafana is an open source piece of solution that is available to anybody. They can download it for free. Rockwell understands where they sit, especially when it comes to the dashboarding game. And it's not necessarily smart for us to try and jump down the road of developing these large bombastic dashboarding solutions, and there's already so many good ones out there. That a lot of customers are already taking advantage of. So what we want to do is invest in the right areas and invest in areas that can greatly impact our customers. DataMosaix was obviously one of those. Now Power BI is still widely available for people in the Microsoft communities. And most people who have the 365 subscription actually have access to that. Grafana is available commercially, and it's an open source solution and most SI's absolutely host that for them.

Thank you very much. So we are approaching time at the end of the webinar. [Operator Instructions] So there is this question, can I deploy this on my companies as your tenant or our private cloud?

Not at this point in time, no. Not initially, DataMosaix is supposed to be a SaaS application hosted by Rockwell. But I want to explain one of the reasons why, right? We are hosting it in Azure -- in an Azure space, and you'll get [ graded ] towards your Azure commitments and so on. We're investigating the private cloud feature. But the thing is the amount of firepower and now horsepower that goes into doing something like artificial intelligence and entity matching like that. It's a mix, and it's something that it becomes a barrier for people to utilize this tool set. The cloud age is coming and most people are going to need to either learn to get on board or at least work with it because it is the way most places are moving for one, but for another, you're gaining so much benefit by allowing that within your business.

Great. And is this similar to Domo and Tableau. Is there a list for extractors...

In terms of Domo and Tableau, I know what those are. And it's so much more. So Tableau has a tool set to be able to visualize information. It's fairly good at being able to dashboarding solution. Becoming a dashboarding solution as well. But what it lacks is that fundamental thing that DataMosaix has, which is a huge and rich capability of data contextualization and organization. The asset hierarchies, the file management, the data management and data pipelining capabilities and as well as the extractors really are what set this apart. Now Tableau, it could be something, especially if your business uses them on a daily basis as the output right? And it could be tacked on to the end of DataMosaix where DataMosaix is the thing that's organizing your information and Tableau is the thing that is visualizing it and bringing it out to the [indiscernible]. And in terms of a list of extractor, I believe you can find those on either Cognite or our website as well.

Sounds great. Okay. So we have time just for final question. So this will be the final question to submit. [Operator Instructions] The final question that I will be asking is, what is the cost structure, a question by John.

Yes. Great question, John. So it is going to be basically 2 different subscription solutions, which it is going to be an annual subscription. And you're going to have basically a standard and an enterprise solution. I don't necessarily -- I can't speak to the pricing because locally Canadian here, but there's a lot of Americans, we're not going to give in the Canadian rubles. But at the end of the day, the pricing structure is on the portal -- it can be ordered by the portal. It can -- I highly recommend you reach out to a Rockwell person to see what specifically works. But fundamentally, we're going to start with that standard subscription model, which is 1,000x [ curious ] data sources. It's limited to 2 specific applications in addition to Rockwell supplied applications. There's a little bit of nuance to these, and then you have an enterprise subscription as well that really gives you the whole [indiscernible] everything from standard [ tier ] unlimited applications, 3D file support, multisite implementation. Again, when we're selling something like DataMosaix, we really have to understand what is the value it's going to bring to the customer and ensure that, that value is there. And it all comes down to talking with your customers, being tuned with where they're at and whether they're ready for something like this. And I posit that so many people are ready for something like this. They just need to be -- they need to take that leap and go through it because the benefit will far outweigh the effort down the road.

Perfect. Thank you very much, Jared, and Chris, for this perfect presentation and for all of your questions. So don't hesitate to reach out anyways. With that, I would like to thank you for attending this webinar. And in an effort to improve the webinars and provide topics of value to you. We kindly ask for your participation in our brief survey that will pop out right after you close the webinar platform. And if you'd like to speak to a representative for more information, you can make that request in this post webinar survey. So we look forward to see you again at our next event. Thank you very much, and have a great day. Bye, bye gentlemen,

Thank you so much. Have a great day to you.