Why a Federated Data Team is Crucial for Business Value, with Dow

The Data Chief is a podcast for data and analytics leaders to share their personal stories and insights on technology, culture, and leadership. Welcome to the Data Chief. The Data Chief is brought to you by Thoughtsbot, the AI-powered analytics company. With Thoughtsbot's user-friendly platform, anyone can leverage natural language search and generative AI to uncover actionable data insights. Whether you're a seasoned data analyst, a business leader or product builder, Thoughtsbot empowers you to effortlessly interact with live cloud data, generating granular, hyper-personalized insights in real time. Learn why customers at SIGNA just eat takeaway and capital one trust Thoughtsbot to drive business outcomes with data. Visit Thoughtsbot.com today. I'm your host, Cindy Howeson, and I'm excited to introduce two guests from Dow, a company at the intersection of material science and digital innovation. Joining us today are Chris Brumman, Dow's Chief Data and Analytics Officer, and Dan Futter, Dow's Chief Commercial Officer. As many organizations are trying to balance that centralized versus decentralized data mesh operating principles, Chris and Dan are true partners in uniting data and business. From Dow's integrated data hub to their visionary data and AI literacy programs, the company is transforming how data is accessed, understood, and used across the organization. Today, we'll dive into these initiatives and how they're paving the way for sustainability, innovation, and global collaboration. Chris and Dan, welcome to the Data Chief. Thank you, Cindy. Thanks, Cindy. Great to be here. Yeah, so where is here? Chris, where are you joining us from? I am in my office in Midland, Michigan, which, Cindy, I know you're a little familiar with, and I will tell you, the sun has made a rare appearance in January, which you know is unusual here. It is unusual, but you're making me long for the tributary that where the three bridges come together, and of course the Santa House. We're Santa's from around the world come to train. Yeah, it's quite an event when they all show up and down. Yes, and Dan, what about you? Where are you joining us from today? Well, I should be with Chris in Midland, but the last two weeks it's been in the minus F's, you know, very, very low numbers. So I decided to base myself in the UK. So I'm actually in my second house in West Sussex, just south of London. Okay, great. Our London listeners will love hearing that you are joining them locally from there. So Chris and Dan, you have both been at Dow and part of this data in AI journey for a long time. Chris, why don't we start with you? Tell us a little bit about your role and how you came into actually in one of the first world CDAOs, so the A is part of your job. Tell us a little bit about that. Yeah, thanks, Cindy. So if I describe the role, I always talk about it in two buckets. One is of course leading our centralized team, making sure we have organizational strategy. We're lining up what the business wants to do. We're attracting and retaining the best talent. So leading that team is a big part of my role, but then I also think of it as there's an enterprise side of it to make sure that the company is doing what it needs to do with data, analytics and AI. So I think of it in both of those. And in terms of how I got here, my journey into this role started just under three years ago now. I was leading what we call enterprise systems, which I describe as ERP and friends. And we had an enterprise data team within that organization. And what we decided to do is take that data team out, combine it with our analytics team that was sitting separately and put them together. And that's what's when the CDAO role came together. And over the last three years, I think we've seen the power having the data team, our analytics and AI experts all in one organization working with the clients across the enterprise in a kind of collaborative manner. And it's funny because the timing was interesting. We did this again early 2022. The AI hype was already starting a bit. But then of course later that year, check GPT hit. And for anyone in my role, the hype of AI just dominated and to have that organization in place with the combined data analytics pieces made a lot of sense for us. Yeah, for sure. And I think it's interesting that your background was from the ERP side. So clearly SAP as a big ERP figures into that. And oftentimes data is an afterthought and thought of as just an exhaust from ERP systems or digital interactions. How do you think having had that background, it has impacted how you view data for business? Yeah, that's a good question. I guess I would go to, if you think about the ERP space and just that enterprise space in general, everything is around business processes. And so one of the things we talk about is and then business processes because you can't just work just in supply chain. If you want to look at an end, you need to know from an innovate, to marketing, to producing, to selling. And so when we think about data that way, we think about data really aligning into those business processes. So we talk about end and business processes, but then we talk about end and data threads to go through that. So I think having been in that space and understanding the business process aspect of it, it served me really well in terms of making sure that we start with the data. I know everyone wants to talk about AI and it's exciting. It's a sexy part to talk about. But being grounded in the importance of the data, I think it's been important for me. For sure. And we know you can't do AI without data. So Dan, let's hear from you. How did you arrive at this role within DAW? Yeah, so my passion is all about customer experience, Cindy. So maybe not as normal in the B2B space, a big industrial company like DAW, but we're very large, as you know, and we are dealing almost every geographical market and almost in every vertical market you can think about. And so it's quite difficult for us to create one coherent experience of what it's like to do business with DAW. And so I get heavily involved in trying to make that come to reality. And clearly, and this is a bit where Chris and I overlap so much, some of those interactions with us are digital. Some of them are through people and there's lots and lots of on and off ramps between those two worlds. So it's trying to make sure that those on and off ramps are also intuitive that customers get as easy, enjoyable, and effective experience, whatever route they're coming into us, whatever job they're trying to get done. So I find myself involved in all aspects of the commercial face of DAW, but that probably is the simplest crystallization of what it's about. Yeah, that's great. So if you can be more specific, Dan, like paint a picture in the B2B world, how data is used, you say it's customer experience, are your customers thinking about on time orders, best pricing, product quality? Can you refine that a little bit? Yeah, it's all of the bar, but I think that I would break it down into three. Three spaces. I mean, the first one is what inspires somebody to come and look at us or one of our products in the first place. And so a lot of our interaction with customers, how we inspire them, whether it's direct outreach or whether it's more passively what we put out in to the public domain that they hit on, is designed, of course, to tell them some story that's relevant to their market. And make them want to click through and see more about our products. And so, you know, our second job that we get done is really helping customers choose our products. At any point in time, we have between 15 and 20,000 products. So it's hard for customers to navigate to exactly the type of information that they need to be able to make a decision. Yes, this is in the right ballpark. I'll line out to talk with somebody now. And so how we manage our catalog, how data is presented from that catalog, how other textual information is surfaced based on how customers are looking for solutions is critical. And then the third piece, once you've selected a product, the order experience and you touched on a couple of examples of what pricing I'm going to see, what lead times can you offer me. Can I then track a product once I've ordered it? Can I then manage my invoicing online? So the whole through to sale transaction process and then post sale experience, trying to provide as much of that optionally for customers if they prefer to interact with the through digital means is part of our battle. And so they're back to what Chris was saying about quality of data and processes generating high integrity data because it's critical for us if we surface something about our products or we surface lead time or we surface information. And when you're orders coming, that is accurate. And so again, this is where Chris, it seems odd, right? I've received commercial officer to be talking so much to the data side of the business, but it happens all the time because that is digital. It's enabling those experiences for our customers. Yeah. Well, and Dan, you say it seems odd for you and Chris to be talking so much, but should it be odd? If the way we delight customers and know the customers and keep them happy is increasingly through data, maybe it shouldn't be so odd. Yeah, I agree with that. I mean, I say odd because maybe if you've asked me 20 years ago, I'd say that was odd. But if you asked me now, if you're not doing that, I suspect you're going out of business because, you know, all of us, whether it's in our private lives or our work lives, you know, our first port of call, if we want to find information on something, we want to place an order for something is, can I do it digitally? I don't want to have to pick up a phone. I don't want to send an email except in extremist, right? So we're trying to create that same, you know, private life type digital experience for our customers that they can also enjoy at work. Yeah. And so, yeah, you say 20 years ago, it would have been so odd. I do think back. I think most people know I started my career in the space at Dow 30 years ago. And, but I started in the business. And when I went over to IT to work on the SDN, the business who had been my friends were much more wary of me at that point in time. So, so Chris, let's go back to you. Some people use the term data mesh at Dow. You call it hub and spokes. Tell us a little bit about why you're adopting this approach. Yeah, certainly what we're trying to do is based on the data mesh concept. It's become very popular in our space the past few years. If I talk to date, if I talk data mesh to my business clients are going to be a blank spare rights. Now we use hub and spoke. It's the visual makes a lot more sense. At the end of the day, it's really about decentralization. I think that is the key piece of this, right? And so why is that important? First of all, you started to see over the last few years just a growing demand in this space, whether that's data governance and data quality efforts. It could be reporting in dashboards. It could be AI and ML models. That demand just keeps growing and trying to funnel that through a centralized team. It's just not going to work. It will become a bottleneck. It's too slow. People get frustrated. And so by putting this hub and spoke model in place, we still have, I say the benefits of a centralized center of excellence. We provide expertise and consulting for the company, guard rails and policies, things like responsible AI framework, support for whatever. And certainly a big piece is the data platforms to get that data accessibility. And our team still does a fair amount of design and build work where we need to. But that's the role of the hub. But then having the spokes, so the businesses and the functions that then can bring the combination of their domain knowledge they have about what they're doing every day and the analytics, the capabilities that we're giving them. It's just a lot faster. It's more effective for the company. To be honest, it leads to people from a career standpoint. A lot of people want to learn this digital side, the data side. So that's kind of another benefit we're seeing for people. So it just, it allows the speed, it allows accuracy, it allows effectiveness that we just can't get one, have a super centralized model. Yeah. And so you list some of the many benefits and it does sound like it's getting that balance between economies of scale, the guard rails, the policies, but then the agility that the business needs, the speed to market. Dan, how has this impacted your ability to still get to trusted data, but maybe business specific data faster? Yeah, I think that probably what started Chris and I off on this journey was thinking about what would it take to confirm a customer's order real time? So if they come into doubt.com in our case and they try to order, can you give them an order confirmation in the moment? Can you then give them a delivery date in the moment? Now it seems really easy, right? We live that in our daily lives with Amazon all the time. That had never been done in the industrial space. And so that was our first sort of step into that. And he was thinking about, okay, what data would be required and what level of integrity would that data have to be able to surface and commit an order in the moment a customer, customer places it? And then you think about that, it starts to spread across to all the different touch point experiences that customer would have. So what is my pricing? How do we make sure that accurate pricing comes in front of the customer? How do you make sure that accurate lead times coming in front of the customer? How do I make sure that accurate invoicing is represented to the customer? All of them in the end have a come on foundation back to that hub and spoke idea, which is, you know, Chris, I suppose is agnostic, we have to say to him, here's the type of data we would like to see because here's the type of insights that we want to have or the type of experience we want to create for a customer. And then it's that back and forth to say, all right, where is that data? How does it get created? How do we ensure that it's high integrity, which processes generate it? Which people are engaged in those processes? How do we make sure we train them so that abinisio, it stays, it stays high integrity? So it started with that journey and then you realize it just blossoms from there into lots and lots of the other day to day insights day to day jobs that we're trying to get done. Yeah, and so it's still about the process and what is the customer, what is the customer journey? How, how well do you think it's worked, Dan? You, you have centralized data and data that is captured internally and then sometimes you have data that maybe only you, your team needs or your function needs. Is that something that Chris's team still handles or do you have the guard rails and the policies to pursue it on your own? No, typically still work with Chris because it's very rare that we have an example where we just pull in up a single data point and generating insights and then action from that. Typically, we want a more comprehensive view of something. So if I give you a given example, maybe it is in how we surface information onto our website based on a customer search. So historically that was straightforward but not that helpful for the customer, though, is they would see a data sheet, be able to click on it or see some marketing collateral, be able to click on it. But then they would have to hunt for the insight that they were looking for within those documents. And what's changing is now based on the type of search that a customer might make, we could call up specific text from within documents, specific data from within documents and curate that probably in the right language for that customer so that you actually explicitly answer the question. You don't just get them to the right street and then expect them to get to the front door, but you get them through the front door just by surfacing the right level of information. Now that's coming from data sources, that's coming from textual sources and it's coming from different databases, all these different data origins. And so that's where we need Chris's help. So how do you do that real time? How do you structure that so that you can come up with those comprehensive answers, maybe sourcing from five or six different places at once in a curated fashion? Right. So you understand the ultimate end question, ultimate point of impact. Chris understands how to make the technology work and where all these points come together. So what are some lessons learned? Like Chris, what's been your experience partnering with the business and keep in mind a lot of CDOs and I'm using the term specifically CDOs, they can be a little shy of the business. So what are some lessons learned that others should take away from this journey? Yes, it's a good point. I would say you talk a lot about this on your podcast and out on LinkedIn, Cindy is the first thing at a high level is just change, move into a model like this, it takes longer than we think, right? It's a challenge. I was probably overly optimistic when I took the role of how quickly we could move everybody into this different type of operating model and how fast we could implement the integrated data hub to make that data available. It's tough going through that kind of change. I think another lesson learned, you just hit on it, right? Is one of the first things I did when I took the role is went and sat down with, let's say the top 15 leaders in the company and spent 30 minutes with them and said, okay, what is important for you in this space? How can we support your strategy? Whether it's a functional or business strategy? You have to know what the company wants you to do and I think some people just jump in and go and you need to take that time to understand where everybody's at. Beyond that, I think a couple other things lessons learned, one is trying to define specifically what you want to have happen, let's say in the hub and what you want to have happen out there in the businesses and functions and then realizing that's going to be different for every part of the company. So what's good for one function or business isn't going to be good for another. So while you have this structure in place, you have to have some flexibility to it and be able to kind of move as they move. I think that also providing the upskilling support so that those clients can build the skills they need, they gain confidence in the state and analytic space and then start to demonstrate the benefits of what they're doing. So we're trying to just make life easier for those folks. let them be more effectively. And at the end of the day, provide more value for the company. And Cindy, you've been hitting this value driven topic big time lately, right? It's everything we do needs to help provide bottom line value for Dow. And if we can help the clients out there, get to answers more quickly through the use of analytics, then we're helping with that and just bringing as much value as we can to that bottom line. But I think if I go back to the number one lesson learned, it's the relationships across the company that we've talked about in understanding what the different parts of the company are trying to achieve. Yeah, that's great. And so your first, let's say, your first 90 days, meeting with those 30 different stakeholders. So Dan, when Chris first called you, what was your reaction? I think that probably there's a nervousness, first of all, is I've got to try to understand what Chris is talking about, right? So he says an integrated data. Well, what is that? I don't know. I don't use that every day. So I think sometimes it's that you've got to get over the translation issues that we're speaking different languages. But we've been working at it so long now that we've got a really, I think, naturally fluent way of working with one another. And he's asking me the right questions, which is what are you trying to get done? An example we had quite recently is, well, I'm trying to predict better what our customers want to order when. Which sounds really easy. But when you got that many products I mentioned earlier, and you've been sold in all the markets around the world, and you've got longer or shorter lead times, depending on where you're shipping to, it actually becomes really critical to have right, product, right, place, right time, right? It's hard to do. And the work with Chris was really good, because he says, well, you just try to put it in your terms. Like, well, I need to know what customers will want when. And so then you start to talk about, well, how could you do that? What type of tools could we use externally that would help us get a better idea of forecasting? What indices are out there? How do indices relate to one another? What algorithms might be able to better predict for us so that we could do a better job of that, right, product, right, place, right time? And so Chris has got that great ability, I think, to ask me the right question, here would I put in our terms, like what a seller might say, what a marketer might say, what a customer might say, and then translate that into, well, if you're going to try to do that, then you need to be able to access these types of data sources in this type of way. And I mean to structure things in a way to make that easy. So you're doing that over and over again. And I'm the lucky end, so I just have to say what I'm trying to do. And Chris has the difficulty in which he's trying to translate that into, this is what the data and the text will have to do. Yeah, and go ahead. I'm sorry, I was going to jump in. I love the point in dance making. And you know, if you talk about some of the foundational keys to success in the space, what dance talking about is data analytics literacy, right? And going, you know, so one of the first things we did back in 2022 is work to stand up a data literacy program. We worked actually pretty close with Val Logan, who I heard on one of your podcasts late last year. Valerie refers to us as the Dow 17 from the cohort that we sent through. And in that cohort, we took a person from commercial and dance organization. We took a person from manufacturing, a person from purchasing, a person from each of our business units. Because to dance point, if we aren't all able to speak some common language, then we're going to be talking past each other, right? And it's a really good acknowledgement that Dan gave, because like I said, if I go out and talk about data mesh to somebody outside of the tech area, they're going, "What are you talking about?" So part of this is trying to make sure that we all have at least enough understanding of the basics. And then on the IT side, making sure we don't talk in tech jargon all the time. I think we like to do that, but if we do, we're not making sense to anybody. So I just want to jump in. I think this focus on data literacy early and what you're trying to do is just super critical. Yeah, and thanks for giving Val a shout out, both a dear friend and former colleague at Gartner's. I call her the Fairy Godmother of the Data Literacy Movement. And so happy to hear that was part of the early program. I do think though, or let me ask a question, I'm already leading the witness the way I'm framing this. So we need more data literacy, but do you also think we need more business literacy? Do the data professionals need more business literacy? Absolutely. If I think about, even our 20, 25 objectives, of course, we're at that time of year where everybody's rolling out their goals for the year, at the very top of our organization, if you were to ask our CIO, we would say that's one of the top things we need to focus on. Not just in the data space, but really for anyone in the IT and technology space is business acumen. And I'm not talking financial acumen. I'm talking, you know, how does Dow consumer solutions make money? Right? How do we go to market? How does plastics do that? We, I would say historically haven't spent enough time on that. And again, I think what you're hitting on, as we're all working more and more closely together, it's not like the old days where there's this IT function sitting over here and the business sitting over here, we're having to talk every single day to get things done for the company, digital is part of almost everything we do it down now. So getting our IT folks, our data analytics folks, to have a much stronger business acumen is actually a big focus for us this year. Yeah, great. No, I think of Bill Schmarzo, the dean of big data. He talks about how it's either how do you make money or what costs you money in trying to make money. We have to understand those two sides. You know, with the hub and spoke model, one of and the need for agility, we talk about co-location. How has that worked? Dow is so global. You have your pockets of different centers of gravity, let's say. But what are your thoughts on co-locating the spokes of the data and insights team? Does that matter or not really? Are you asking, Cindy, co-locating of the actual data itself? No, of the people, the people building the models, pulling the data sets together, designing and deciding the outcomes and the priorities. Yeah, I'll give an example on and dampry us in thoughts here too. One of the structures actually we put in place quite a while ago now is, we call it diamond system solutions, but that doesn't matter as much as the concept is back to this kind of business process space. So if you think about, I'll use order to cash as an example, right? It's a process everybody understands. So we have order to cash business process experts, let's say that sit in the supply team organization that kind of bleeds over to commercials well, but they sit on the functional side so they know how the functional work gets done. We have order to cash IT team and data folks that work on the actual data sets there, and they all sit on kind of one virtual team. So there's an order to cash team that consists of process people, IT people, data folks, and although they report in separately, let's say to supply team and IT, they are a virtual team that meets on a regular basis. You also have that for manufacturing, purchasing, commercial, etc. So we try to do that. It is more difficult from a physical location standpoint, as you said, we have people all over the world. But I would say there, one thing that we've obviously learned over the last several years is how to do more of this work virtually with the use of the tools that we have. So a lot of it does happen that way, but then certainly even in our centers, we have five or six kind of major centers around the world. We do try to make sure those folks physically are sitting as close as possible just to enable that team interaction. Right. Thank you. Yeah, that's helpful. You both have referred to the Integrated Data Hub, which came out in 2022. And in 2024, CIO Magazine gave it an award for IT leadership and innovation. Can you provide some callouts as to how it supports the fundamentals of easier access to data? Like was it, was it so hard before what had changed? Yeah, I can talk a little bit about that. First of all, thanks for bringing it up. It's, it was a great honor for our entire team of Dow, right? Kind of definitely a team effort there. In terms of how the supports are efforts, I go back to maybe a story from when I, when I first took the role, is as I'll talk into folks, both on our team and out kind of in our clients. And this isn't necessarily the senior leadership, but talking to folks that are practitioners. One of the things I started hearing over and over is that if I'm a data scientist, and I have a business challenge, I'm trying to address. The minute I get ready to go, I kind of pause and I spend months. Sometimes three or four months I was hearing to figure out who owns all the different data objects I need access to. How do I get access? How do I get them to one place I can get them? How do I get security access to those data sets? So that's obviously extremely ineffective and very, very frustrating for our data scientists and our data analyst. So as, as I was talking to the team, you know, what I kept hearing is we need data science in a day. That, that's kind of the vision, right? As data science in a day versus months. So to do that, we started standing up this integrated data hub. It's a Microsoft stack and we started put. data from across the company into that one platform. So again, back to the thinking of things end to end from a process and data standpoint, you know, we have marketing data, manufacturing data, supply chain data, purchasing data, etc. All going in there. And we prioritize that based on actual use cases coming at us. So not just randomly putting it in there, but take the top priorities for 2023. Let's get those data sets in there and then we'll build over time. I would say getting there has been no small task. I think about some of the challenges is things that might seem simple, like even where kind of high level three qualifications before we put something into this new platform. We needed a data owner, some level of metadata, and some level of data governance and data quality. Seems pretty simple. It's kind of amazing how hard it's been to get through that. Just this is a big company, right? So who owns that data set? Who's going to take responsibility for the metadata and the data quality? But we kind of stuck to our guns and said, we do not want to put garbage into this new hub. You know, the garbage out mindset. So we've been really good about, I'd be kind of sticklers on what goes in there, which means it's a little slower than maybe everyone wants, but I'm very confident talking to other CDOs and CDOs. If you don't do that, you end up with a huge tech debt issue on your hands as soon as you get this thing live. So I would say in addition to the data science and the day, the other thing that this data hub has become is kind of the go to spot for any of the new capabilities that we're building. So if you're building a sustainability application or supply chain disruption, what they all do now is they're kind of looking at that data hub to be their source of data versus going to six different places and trying to plug that together. They can just come to one spot now and get that. That's been a huge win for us. I think back to we do this data science challenge every two years. It's a big event for our data scientists and even our business folks get involved. Dan was the co-sponsor with me last year because our theme was growth, but it was interesting. We do like a shark panel at the end of the event. We all get together and Dan and I were sharks and it was interesting watching the teams come through and present their ideas on how to tackle growth. They would inevitably talk about scale and they say, of course, we would get the data from the integrated data hub. I was just kind of sat there as one of the sharks going two years ago. No one would even talk about this. Now it's kind of a key predecessor to all of these ideas being successful. That was kind of an aha moment for me. Sometimes you get so close to what you're working on. You don't realize maybe the progress that's being made and all of a sudden you see a day of people coming through and saying, for me to be successful, I need that data hub in place. Yeah. Well, so there's a number of best practices in there. So a couple things. Let's hear Dan from your perspective. So first off, do you have data scientists and insight analysts reporting directly to you or is that all funded through Chris's team? I do have them reporting directly to me. So I'll give you a couple of examples in a moment, Cindy. I'm very glad by the way that Chris answered the integrated data hub question because one of the benefits I have in my job is I don't need to know what that is or where it is or what's in it. So our conversations tend to stay at the, what are you trying to get done? And then the Chris has to figure out how to surface the data to do it. But so I was learning actually as he was answering your question. But anyway, back to the data scientist question, two areas where we specifically have data scientists working with us in my group. One is on pricing because pricing is so complicated. I mentioned that many products, that many interactions, those products with customers, that many different pricing conditions are the spot or contractual. It's super easy to get it wrong. So we're constantly doing analysis on our pricing patterns. And the second area is around the whole field of churn, which is more complicated actually than the pricing. The pricing tends to mostly originate in our own ERP systems through things that we've done in the commercial organization. But on churn, it could be looking at customer buying behavior, could be looking at what they've said in a survey, could be looking at what they said in the last visit you had. It could be something they've posted online. There are so many different sources of information that might inform you that a customer is, you know, thinking of increasing their business with you, you know, thinking of leaving you for a competitor, that that's one where we're building out how the data has to be structured and has to be surfaced to help us understand and predict when we might see, you know, a customer defecting to a competitor so that we can cut that off at the past early and intervene. So both of those areas, it's been good for us to have data scientists actually in our spoke to be able to work with us on the details. Now those guys find their community of practice within Chris's world, within Chris's teams, but they're actually seated within our teams. There's a little bit of an example of the co-location you talked about and we they're paid for directly by us. Yeah, well, I think it's that domain expertise, but then also the skills and career progression and it's where do they want to play. Now, so if you think about before and after Chris described how it could take months for a data scientist to fetch data from different places, where was it? Now you have the integrated data hub. What's been the impact on your team's ability to ask these questions? Oh, I think it's had a huge impact. We just take that churn example. There was no way two years ago that we could have interrogated all those different sources of information, either textual or data. And even the data ones, there might have been housed in different places, structured in different ways that we couldn't surface them in comparable ways. So I just don't think it was possible for us two years ago. So the difference is we can do it now and actually you can do it quite quickly with the help of AI. Whereas two years ago you just couldn't. That would have been a really, really intense and long interrogative exercise. Yeah. And so you have your priorities, you have your data, but you also compete for resources across, I think Chris, you said 33 different functions and business units. How do you, let's say produce product comes along and says, well, we have something that's a higher priority than this churn analysis. How do you reconcile the competing priorities in this Hubbins-Boke model? That's a good question. I mean, that's really true of any space in the IT and digital. There's always more demand than there is supply. I would have said several years ago, Cindy, it would have been a bigger issue because things were so centralized. To be honest, like manufacturing, for example, research and development, supply chain, dance group now, they have a lot of data science and data, in some areas data engineering skills, even mostly data science. So it's become a lot less of an issue. Now certainly, there's still kind of a funnel of work coming in. And we have prioritization processes. We put a lot of time into that, let's a couple years. At the end of the day, it comes down to most often, which opportunities are bringing the most profit for Dow. Obviously, there's more than profit. There's safety and employee experience and things. But at the end of the day, we have to be really tough on a project. Maybe very exciting. It sounds like a great thing to do. Are we pretty certain we're going to see an impact either growth or productivity in the company? If not, why are we working on it? So it sounds simple. But at the end of the day, we try to put financial value against every idea that comes in. One of the things I would say we're trying to do better now is have a pretty senior level sponsor that's going to sign off before we start the project. I always joke with folks. If I go to external CDO, CDO forums, this is one of the top topics. We're all trying to figure out how to show the company the value that we're providing. And I think too often to wait until the project's done or the capability is built. Then we try to figure out how to get the value and who's going to sponsor it. We have to flip that around and say, okay, before I even start working on this project, who's the senior leader who's going to sign up? Not just a name, but be engaged. Be really part of what we're going to go do. So that when we do complete it, we can look back and go, okay, now we know how to measure the value. And there's tangible financial value to what we're doing. Then take the idea to reinvest that into the next project. But it's a difficult area. This value tracking, value management, value sponsorship is a challenging area, but I think the most important piece is just getting that up front. And then you know what to prioritize and what to work on. Yeah, absolutely. We would call it business value assessment. And I've always said, even if it's just a back of the envelope calculation, I know, I know Dow is more rigorous. I'm flashing back to. We had the economic evaluators put us through the function. What's the payback period? What's the ROI? What's the discounted cash flow? We are a copy of engineers. Yes, I know. That's still there, Cindy. It still exists. But now I would say business value assessment, do the back of the envelope calculation. At least, ideally do it more rigorously. And then business value realization. I want to go back to one of the best practices that you shared. You do an annual hackathon. And I love the Shark Tank example. But you also, I understand, have an internal podcast. to drive change and ensure alignment. Can you share a little bit more about that? - Yeah, actually, Dan, I wanted to come on. He has a podcast too, and they talked digital on it quite a bit, but it was one of the things a lot of my team members and kind of people in our community kept bringing up after I took the role as Chris, you ought to have a podcast or so much excitement about data and analytics and AI. So once I got settled in the role, I didn't launch that. And over the last 18 months, I've brought in a lot of the senior leaders around the company. I really wanted to start with not digging into the technology on that podcast, but digging into the business reasons that we're doing, the work that we're doing. So, Dan was a guest on it. I had leaders from different businesses and functions join. We had kind of an external guest wants joined to help talk about what other companies are doing. And it's been a pretty big hit. It's a big learning for me, Cindy. You're an experienced podcaster, but there's a bit of a learning to go through to post one of those things and keep the conversation going. But the whole point was to get easy access for any employee that was interested to listen to what senior leaders are thinking about data, analytics, and AI. I would say we're making a shift this year actually. We're relaunching the podcast coming up. And I'm going to try to shift from just having the more senior leaders on to having more practitioners and folks that are doing some of the work. So yeah, it's been fine and exciting to do. And I think it's been a really good way of learning for people around the company. For sure. Dan, tell us about your podcast. It's been going. It literally started off the first week of lockdown of COVID. And by chance, although it ended up being fantastic timing for us because, you know, what we'd been seeing is actually I got the idea from my daughter, strangely enough. It just started work at Fritale. And then she was out with a driver doing deliveries of chip hacks. She said, so how do you get your company information? Right? You read emails and they're like, I'm driving all day. I don't read emails. And they said, but I know what I would do. I would listen to a podcast. So they invented this thing called chip chat, which I thought was great bit of a iteration. And she was telling me about this idea just before COVID was not a great idea. I bet it's the same for our folks, right? They've all got different ways to absorb information, some by email, some by webcast, some by, you know, whatever, phone calls. And we have never really offered podcasts as an option. So we started one then. We've run, we'll make, I think, about 100th this year. It's just used internal. Any subject you can imagine we think might be interesting for the commercial function. And it was the first internal podcast that down. Now we've got a few of Chris's talk through his, which is great, great fun. I was on it. And I suppose what it does is it's a little bit like you, you know, obviously you've succeeded. Here's make something conversational and much more accessible to people and hopefully more interesting. And maybe meet the need of lots of different ways of learning about things and different ways of exercising your curiosity. So it's actually really nice to be this side of the microphone. Cindy, nice, nice podcast is very small compared to yours, but rather than the other side. But we've found it to be super popular way for people to add to the way they find out about things. Yeah, it's letting them consume the content where they need it. If they're in commercial, they're driving around, flying around, whatever. So, and you do have a lot of employees in doubt. So I don't know. It's a big stage, damn, in Chris. Well, I feel like-- It is, it's not yours, but I'll take it. OK, there you go. Well, now you're on this stage too. So it's a bit of both. And I-- so I feel like we could go on and on. And I want to ask you about people that are still there, technologies. I feel like I have to just show you this one thing that Mark Carter had sent me. It was the chips from the original Dow S-D-N that you decommissioned only a few-- Well, that's great. I know. I felt so privileged. He sent me his slew of chips there. But so, definitely Dow has a place in my career and in my heart. But let's go ahead and pivot to a lightning round. So Chris, who do you find to be an inspirational leader in the tech space right now? Yeah, so this one's interesting. To me, when I think about this, there's an obvious answer. And I will say-- I mean, because of his entry in a politics, this might be a polarizing figure at this point. But I can't think about that question. I think of Elon Musk. And the way I think about it is, if it were just Tesla and what he's done with kind of redefining how we think about electric vehicles and the mean performance cars, and what he's trying to do with autonomous vehicles and who knows what becomes of RoboTaxi, if it were just Tesla and be impressive, then I go, OK, if it were just SpaceX, and what's happening there. And I think back to last fall in that video of them catching this super heavy booster, he's always got these funny names for things, right? I watch that video. I was like, that's just unbelievable. So we're just SpaceX. And then I could say, what if it was just XAI and GROC is kind of coming up with the big LLMs now? And if there's any one of those three, I'd say it's pretty impressive. But to be leading all three of those technologies, it's pretty impressive. I said, politics aside, and your opinion of them in that front, I think, is a technology leader. He's a kind of a generational figure. Yeah, he's fascinating. Dan, what about a book or a podcast you think everyone should read or listen to? Ooh, so I have a podcast. It's not on tech. Occasionally it's on tech, but it's not on tech normally. And I know you now have a British husband. So it's actually one he may know. It's called In Our Time. In Our Time. And it's about 1,000 podcasts in just to give you some sense of how long it's been going. And it's hosted by a guy called Melvin Bragg's Long Time Journalist. And it covers every topic imaginable. But it does it like a 45 minute slot. It does culture, it does history, it does art, does science. And that is my favorite walk in the dog podcast. I love that podcast. And with 1,000 of them, I don't think you can ever get through all of them. No, you can't. It would take years. How about Chris, a song or an activity that pumps you up after a hard day? So a full confession, Cindy, this is what idea I completely stole from you from my podcast. I remember asking Dan this question. I asked their song, their go to song. And it's always one of the favorite things people love to hear. So for me, when you say a song that pumps you up, which is different than kind of favorite song, "Pumps you up." I'd probably go back to my high school days, and I take Paradise City from Guns and Roses. Just the intro to that song is kind of gets me going. It's OK, let's go. Let's go do whatever I need to do. Oh, and I-- OK. I'm thinking our producer is going, yeah, Chris. All right. Dan, what do you consider your superpower? Or what are you doing for fun when you're not working with business insights? Oh, I'll go with one of my doing for fun. So I am a fly fisherman. So I love fly fishing. That's my-- where I find my zen. And actually, I'm normally in a lot in-- if I'm in my home office in the US, I have lots of pictures of fish behind me. So everybody's used to just like trying to figure out what the fish are. I occasionally change them just to throw people off. But as I'm in this house, no fish unfortunately behind me. OK. Well, you'll have to share a picture. We'll link to it or something. We got that. So that's great. And then I'll let you choose your last question. Either, what are you most grateful for? Maybe beyond, of course, the obvious of health and family. Or something that has totally made you laugh out loud. Tears running down your cheeks. Chris, how about you? Yeah, OK. I'll pick the one I'm grateful for, of course, family and health. Like you said, but beyond that, I'm just going to sound cliche. But I am super grateful for the opportunities that now I've been here more than three decades now. I've tried to not get into years now. Just leave it at decades. And I do want to make a special call out. So I've had one boss for most of the second half of my career, Melanie Kalmar, who you know Cindy. I know you two are data pioneers back in the day at Dow. And she's going to move on to retirement at the end of first quarter. So I'm really grateful for her, her leadership, her friendship over the years. I mean, so many people have now been impacted by her and we're going to miss her very much. I'm just grateful that I could be a part of it while she was here. Yeah. And I'll say, so Melanie definitely has always been an inspiration. And I'm so pleased to see how far she's come. She's too young to retire. What is she doing? She'll love to hear that. And you and I know Melanie, I don't think that she's just going to sit around. So I'm sure she's going to be really busy. Great. Dan, what about you? Oh, I give you a laugh out loud as I'm in West Sussex right now. We're surrounded by fesants. And I have one in particular that's made it home in my yard. And I couldn't figure him out because every morning and every night, he is super noisy. None of you have been fesans, neither. But apparently incredibly noisy. And so I've been trying to figure out where he is for the last few days. And I stepped out this morning to try to figure out where he was. And he's roosting in a tree. That's why I can hear him so noisy right near me. And I think I surprised him when I opened the door. And he fell out. And he sort of made it look like it was a glided landing. But it was obviously it wasn't a glided landing because he landed a little bit heavily. And in that moment, we caught eyes. And I know he knew that I did he'd for an out of the tree. That was my fall over funny moment this morning. love. Thanks for sharing that with us. Chris Dan, thank you so much for being on the Data Chief and all the work you do to inspire our industry. Thanks for having us, Cindy. I really appreciate you to let us tell the doll story, but this has been fun. Yeah, it was a pleasure, Cindy. Thanks for inviting me on. It's different to have a different language being spoken every now and then. I learned a lot. Thanks a lot. Thank you for tuning into another episode of the Data Chief. Want to learn more about today's guest? Recommend a future guest or listen to more of the show? Subscribe to the Data Chief podcast on your favorite podcast player. Connect with Cindy on LinkedIn or give us a shout-on-x at BI Scorecard. The Data Chief is brought to you by Thought Spot. Putting humans at the center of data-driven decision-making doesn't have to be complicated. Thought Spot makes it easy. With Thought Spot, anyone in your organization can intuitively answer their own data questions, find facts, and make better, faster decisions that drive real business value. Learn how companies like PepsiCo, EasyJet, and Electronic Arts have sparked their data renaissance at Thought Spot.com.