Building Data Foundations for Agentic AI

[00:00:00] William McKnight: The expectations around AI are very high from executives, maybe from executives that are unaware of the data environment that they’re dealing with and the remediation work that needs to be done there in order to get that ROI. I’m a believer in AI. I think that it is going to kick in those returns that we’re all expecting, but I don’t think it’s going to be anytime soon. I think there’s a lot of work to be done to get to that point. We’ve seen a lot of layoffs in the market. We’re seeing some hire-backs now because the expectations were well beyond what the reality was. But in the short term, I haven’t seen a lot of successful labor reduction as a result of it.

[00:00:42] Scott Kinka: Technology is changing faster than ever, but the most important conversations aren’t about the tools. They’re about the ideas and the people shaping what comes next. On The Bridgecast, I sit down with the leaders driving change from AI and cybersecurity to the infrastructure powering modern business. Together, we explore how technology moves business forward, what they’re seeing, what they’re building, and what it means for the rest of us. I’m Scott Kinka, and this is The Bridgecast, presented by Bridgepointe Technologies. Welcome back to The Bridgecast. Today, we are talking data. It’s never not a hot topic, but one that, in the age and advent of AI, is on the top of everyone’s mind. To have that conversation, I’m joined by William McKnight. He’s the president and founder of McKnight Consulting Group. For decades, William has been advising companies on enterprise data strategy, warehousing, management, governance, all of the above. A recognized author, keynote speaker, and industry commentator who has helped some of the world’s best-known organizations strengthen the way they manage, govern, and, really importantly, activate their data. And we’re looking forward to this discussion. William, welcome to The Bridgecast.

[00:02:01] William McKnight: Thank you, Scott. Great to be here.

[00:02:03] Scott Kinka: Awesome. How did I do on the intro, and what did I miss?

[00:02:06] William McKnight: Oh, you did a great job. Um, I’m gonna have to play that back and use it because it was so good. But yeah, I’ve had a career. One thing maybe that’s interesting in my current workload is that I do a lot of the industry benchmarks. So I’m continually comparing products in the space in terms of performance, ease of use, time to value, total cost of ownership, and things like that. So I got a pretty good handle on that. So, yes, I do work with enterprises, like you mentioned, but I also work with the vendor community as well as industry analysts.

[00:02:43] Scott Kinka: I love that. And we’re gonna get into some of that. I think I have a little bit of data I wanna ask you about from some of the things that you published. Before we get there, just quickly tell us about you. Career, how you got here, you know, how data became the thing. Walk us through, get us to today.

[00:02:59] William McKnight: Yeah. Okay. So, uh, career in data. I love data. I kinda fell into it early. At IBM, they put me on DB2 as an engineer. So I’ve done a lot of different roles around data, consultant. Oh, okay. I went into the enterprise and was vice president of large health care companies’ IT back in the early days of data warehousing. And so I jumped onto that trend real early and built some of the early success stories around data warehousing, and that’s, of course, evolved a lot over the decades. And today, we’re talking about data lakehouses, organized in the data mesh and that sort of thing. It’s evolved, but I’ve stayed with it. I’ve stayed with data and all the evolutions, maybe influenced the industry a little bit here and there, but I’m kind of a generalist

[00:04:50] Scott Kinka: That makes a lot of sense. And I think you intimated in the last answer, but I’ll ask the question anyway. You know, data’s never not been on everyone’s mind, but it’s seemingly sexy again. What’s driving data to be on the top of everyone’s minds?

[00:05:06] William McKnight: Yeah. It’s agentic AI for sure, and agentic AI is taking up a large percentage of budgets these days inside of enterprises, so executives are very interested in it, and they may be going a little bit too gung-ho toward agentic AI without taking a look at the data foundation to make sure that it’s up to a standard where it can support all these wonderful goals of agentic AI that they have. And it’s an absolute requirement because what you’re going to do is send these agents out into the data architecture with MCP or something like that to grab whatever data it needs to do its function. And if it’s grabbing dirty data, if it’s grabbing this data this time and that data the next time, your agentic environment just isn’t going to take off. It’s not going to work well. And many organizations have started down the road of agentic by building agents. Sounds right, doesn’t it? But if you’re building agents on top of a data architecture that doesn’t perform, isn’t governed, doesn’t have security around it, people don’t know where it is, and there’s a lot of duplication, a lot of things that probably have gone wrong inside an enterprise’s data environment over the course of time because that’s how it works, but you have to be at a standard for agentic to work. And so that’s what’s driving a lot of interest in data out there these days.

[00:06:43] Scott Kinka: So you, I mean, you work with big clients, and, obviously, you worked directly with the vendors. So, a little bit of a fun question here. You walk into one of these environments, and you start having this conversation. What’s the thing that you almost always find, maybe that you wish you didn’t? You know? What I want our people to feel better, our listeners to feel better about, maybe where they are right now. What do you find that you feel like you wish you hadn’t?

[00:07:05] William McKnight: I don’t really wish around a client’s environment. But yeah. What I do wish is that what I scoped walking in the door is fairly accurate in terms of meeting the need that they have for what we talked about. But yeah, there are organizations at super low maturity. They haven’t attended to the data. New CIO comes in. What a mess I have. I need help. Okay. Great. Yeah. I think almost everyone out there can be salvaged and kind of get back to competitive level with their business. I’m talking about not just their data. Data’s correlated to your business success. And so I think it’s all more or less salvageable, but one thing that concerns me is if the business isn’t involved in technology decisions, if the technology groups are making all the decisions about how the data should look, we call this governance. If they don’t have good governance, that’s one thing that does concern me, and that’s something that I’m gonna attack pretty early on in a new client. Yeah.

[00:08:15] Scott Kinka: Having the business involved. How do you, you know, the hard part sometimes about that is that you wanna go out and have an outcomes conversation, and then you’re having a data conversation, which is, you know, a precedent, if you will, afraid to get into an outcome what there’s working between. How do you convince the line of business leader that their data is a strategic asset that needs to be tended to rather than just, like, an operational byproduct of the business?

[00:08:44] William McKnight: Yeah. You gotta go straight to the total cost of ownership of data. And that’s the bottom line. That’s really the bottom line. So if you do it one way, and there’s 10 ways to do it, right? But if you do it one way, you have huge carrying costs because you’re constantly going to be attending to that. People are not trusting what you have, so they’re building their own. And there you go. Potential technical debt. Technical debt is accruing in those environments. Whereas, on the other hand, if you approach it with an architectural mindset, you have architectural guideposts. You have a true north for how you want that environment to look today and in five years and in 10 years, according to what you know the business is going to want to do and what the possibilities are. You guide the architecture toward that. And when you do that, you have much better results. People know where the data is. You can really focus on the performance of that data, the quality, the security, all the things that you need to put around the data. So it’s pretty important to take an architectural approach to your data if you want to succeed.

[00:09:59] Scott Kinka: It occurs to me because we went from the technologist to the business leaders, line of business leaders. I appreciate that. Let’s go to the next level. Everything that you just talked about costs money, right? I think, and we’re in this mode, on data, but let’s just talk about it in AI in general. It’s hard to calculate the ROI of either risk, you know, or of doing it the wrong way or of not moving on AI. What nuggets can you give the average CIO who’s out there listening to this right now, who’s struggling with knowing they need to go here and do this work, but they’re not necessarily prepared to walk into the CFO’s office and say that they’re gonna deliver x or y to the bottom line? How do you justify the effort these days?

[00:10:45] William McKnight: Yeah. I mean, you could put it on a probability distribution. You could put risk on a probability distribution, but really, AI is not, at least so far, changing what an organization is chartered to do and what they’re trying to do as an organization. They’re still doing supply chain. They’re still doing procurement. They’re still doing sales. They’re still doing fraud detection. Whatever they do, they’re still doing it, but can AI do it better? Can some measure of, yes, investment in AI make that better? Now I have a rule of thumb. I call it the one-point solution. So if you have key metrics of the business, like fraud, returns, repeat customers, whatever they are, if you can move that needle by a point, not a percent, but a point, like from if you can take fraud down from 5% to 4%, you will more than pay for the environment that it takes to build to do that. So I wanna be sure that everybody on the build team knows why we’re here, what we’re trying to accomplish to the bottom line of the business. Are we trying to improve sales at the end of the day, or are we trying to reduce expenses? And because I speak at the executive level too, that’s what it comes down to. So somebody on that project team, hopefully everybody, but somebody on the project team needs to know why we exist, why we’re doing what we’re doing on a day-in, day-out basis, and that only helps to guide the project toward those metrics and help it be successful.

[00:12:31] Scott Kinka: When you were given a couple of those situations, right, sell more of this, reduce returns, reduce fraud, whatever that is, you didn’t say reduce people. And I know it’s one of the topics that kind of regularly comes up, or at least that’s the fear, maybe unfounded in the market. Have you seen that be a driver yet?

[00:12:51] William McKnight: What I have seen is that the expectations around AI are very high from executives, maybe from executives that are unaware of the data environment that they’re dealing with and the remediation work that needs to be done there in order to get that ROI. I’m a believer in AI. I think that it is going to kick in those returns that we’re all expecting, but I don’t think it’s going to be anytime soon. I think there’s a lot of work to be done to get to that point. And so what we’re seeing is a lot of layoffs in the market. We’re seeing some hire-backs now because the expectations were well beyond what the reality was. And because labor, you know, it’s one of those things that obviously is expensive, that’s something that executives look to cut. I think they look to cut it too quickly, and we’re seeing that play out in the marketplace now. So the future, yeah, AI is going to do something there, and I don’t think we’re ready for it. But in the short term, I haven’t seen a lot of successful labor reduction as a result of it.

[00:14:06] Scott Kinka: Well, it’s so funny. I mean, we have a CX consulting arm in our business, and we’re regularly meeting with the first place. You know? Can we get calls answered with all of this knowledge? And my first answer, getting back to the data, is, well, it depends. Like, how well documented are your processes? Because the people you’re trying to shed are probably the ones who have the intellectual property of the business here to be able to support the customers. Like, what you gotta inform the model on something. Right? What does good look like, and, ultimately, how do you inform it? And then I get the inflection on the other side of, well, yeah, we do have. We have the transcriptions of every recording on a supported call. Well, how well has that data been curated? You know? What makes a call good or bad? How do you wanna train it? So, um, they’re interesting conversations. Help me understand something. You know? I asked you the question early on that was sort of more of the risk question. What’s the thing you run into that maybe you wish you had or what had you? Let’s flip it the other way. You know, in your customers, what does good look like? In other words, you know, who’s doing it right? You know, not by name, just by process. What are they doing differently at the operational level? And, you know, are there best practices that you would wanna share for a business who’s really thinking about getting their data strategy right first before they move into AI? Yeah.

[00:15:21] William McKnight: Um, we walk in the door with a maturity model, and we can plot an enterprise data environment on that maturity model. And it’s informative because we can see where the gaps are to the standard that they need to accomplish their mission as a company. And our maturity model is built based upon the experience of 100 different enterprise clients that we know some measure of detail about their environment. So we build the maturity model based upon not what we sit here and think is maturity, but what actually plays out in the marketplace as maturity. So we look at a company’s growth, their success as a business. Public companies, it’s one thing. Private companies, it’s another thing. A little more complicated, but we can do it. And then we correlate what are the things that they’ve done to data to keep up with that level of success. And surprisingly, maybe, um, they don’t march in lockstep, but there are things that happen on the road to high maturity in a relative order. And so with that in mind, because that’s what enterprises will ultimately do, we look at where they are on that journey across many, many different things, and we look at what’s next and what does it take to get to what’s next? Because that’s when you can start getting into higher levels of return, higher levels of success.

[00:16:55] Scott Kinka: You are listening to The Bridgecast, proudly sponsored by Bridgepointe Technologies. When it comes to customer experience, Bridgepointe delivers innovative solutions that empower your business to stay ahead. Visit bridgepointetechnologies.com to learn how we can transform your customer experience today. You know, without spilling your intellectual property here, which we’re not open to do, but what give me the components of that, like, of a maturity model. If you’re saying, like, step one is somebody’s in front of you. You’re having a cocktail. You know, they’re a tech person. They’re like, you’re in this. I wanna prepare. Like, what are steps one, two, and three at a macro level I should be thinking about as the CIO that perhaps I’m not really thinking about while I’m keeping the lights on in terms of maturing my data strategy to get it ready for, you know, and eventually supporting my AI initiatives?

[00:17:42] William McKnight: Yeah. I would start with good governance, good data governance, and that means that the tech team has stewardship assignments from each of the major subject areas of the business to liaise with in order to make sure that as I in tech am building out that subject area, I build it out correctly for the business, not to what I think it should be in tech, but what the business said should be. So that’s governance, and they come up with data quality rules. They come up with data security rules and all this for them. And then I merely, I say merely, but we implement those rules. So that’s number one. That’s where I would go. Number two, a close number two is data architecture. Having the data architected, not an accidental architecture that you fall into because applications one by one with no thought of everything else that’s gone on before or will come after. Just building applications, and data’s a drag along. Data’s not a drag along. Data is important enough to have its own set of specialists and have its own standards and have its own architecture because you can easily start accruing a whole lot of technical debt if every application does it their way with their tools and with their architecture and their own ideas about what governance is, what quality is, what security is and stewardship and where the ROI is coming from and all this work. So that would be where I would go, is number two: architecture. And for that architecture, I’ll expound a little bit here because, like I said before, everybody’s trying to get into agentic AI, right? We get asked all the time by vendors and the enterprise. So what have you come to in terms of what’s the architecture standard here? What should our architecture look like to support agentic AI? And I don’t know that I or anybody really has the full answer to this because we’re still in the early days on it. Sure. But there is an approach called the data mesh, which we end up with a lot. It’s a nice balance of centralized and decentralized. And what it means is that we build out what we call lakehouses, which are the old warehouse evolved into capturing all data, including unstructured data, for each business area, and I’ll just say that. For each business area, they’ll have their own architecture. It conforms to standards, but it’s their own because they’re the ones that are going to use it primarily. But they share data out to the rest of the organization through what we call data products, which is basically little data integrations where we’ll send that data. The product team might send its product data out to sales, out to marketing, and so on and so forth, in an organized way. So that’s, high level, the architecture that we’re leading organizations into in order to support agentic AI.

[00:21:03] Scott Kinka: Is there, um, you know, I know you’re, I’m asking you these questions in super broad terms, and your answer was perfect. But I’m wondering about the vertical angling on all this. I mean, are the standards the same regardless of where you’re coming from, or are there any spaces you spent time? You had mentioned in health care, like, it’s gotta be a little bit different. Is there a vertical angle to this that changes the strategy generally, or is it just a matter of applying them to the situation?

[00:21:29] William McKnight: I would say not so far. So far, everything I said applies to most all industries. However, of course, the devil’s in the details, and each industry is going to have its own set of data that’s important to its core functions and to its analytics. Right? So health care, you’re talking about claims, and in finance, you’re talking about financial transactions and so on. At some level, data is data, and it needs to be organized into a good architecture for agentic AI. And so the data mesh concept, data products with data governance and all that, that’s all relevant in a lot of industries, in every industry. We work with every industry.

[00:22:10] Scott Kinka: And in that model, when you’re talking about the mesh and then the products, like the LLM consuming a set of data, like, the LLM would be a data product in your mind as well. Do I treat it that way in that structure?

[00:22:21] William McKnight: An LLM would be a place that we can source data from, obviously. And each department would wanna do that differently, right, in its own way. And so it’s really important for organizations today to have what we call LLM ops, which is: which LLM do we use for what, and how do we prompt that LLM, knowing that even if you do that consistently, it changes over the course of time. But companies like, take for example, Walmart. The CIO came out last year, I think it was, a couple of years ago, and said, famously, that without MDM, master data management, without attention to data, with our interface with LLMs, we would be dead because we wouldn’t be able to get product descriptions consistently and rapidly, and you think about all the products that they have. So they do it with LLMs, with good prompt engineering and with attention to their LLM ops. And the rest of us do it as well. But that’s a pretty high use case for LLMs. So that’s a good way to think about LLMs: bringing some data into the organization, into that architecture that I spoke of before.

[00:23:49] Scott Kinka: One thing I hear quite a bit, and, you know, I waffle back and forth, but I’ve even shared the same advice. A lot of people say, where do you start? And, you know, I’ve got some data from InfoTech that says, like, a business is 40% more apt to have their first AI project be turning on a feature in a platform or system that they already have. All right. Simple statement. And if we just think about the proliferation of Office 365 as an example and then the associated SharePoint and things like that, you know, this idea is let’s develop some AI muscle inside the business awareness, right? And let’s just turn it on, right? Hey, we’ve got Copilot. Our files are here. They work with my permissions. Let’s just turn it on. What would you say to somebody who was giving that advice or taking that advice? You know? Just turn on the base AI feature in Office 365 as an example and see what our users do with it.

[00:24:46] William McKnight: On the surface, that sounds a little risky,

[00:24:49] Scott Kinka: Actually. Okay.

[00:24:51] William McKnight: The premise of it, I don’t necessarily disagree with. I like to see things happen in the enterprise and learn from experience and go forward and make sure that, although there will be bumps in the road, we’re moving in the right direction, generally speaking. And you can’t do that if you plan, plan, plan, plan and never execute. You’ve gotta, this is where human judgment remains in the loop of all this, and human judgment remains so important. When do we push on that gas with AI? Where do we do it? You know? What’s easy to do? I think these are the factors in my view. What’s easy to do? What will provide ROI? Because we want it to be important. Maybe not the most important thing, make-or-break this company. You know? Maybe that’s not the first thing, right? But something that is important that provides ROI that we can really watch how we did that and learn from. And thirdly, what sets us up for more down the road? What’s a good preliminary first step into AI that will help us keep going from there?

[00:26:08] Scott Kinka: This has been super interesting. I could talk to you all afternoon. Um, I think our listeners are probably gonna be off their commute by the time we’re done. So what I’m gonna do, I’m gonna move to the rapid-fire segment if that’s okay with you.

[00:26:20] William McKnight: Okay.

[00:26:20] Scott Kinka: So we may touch some topics that we already hit, but I wanna challenge you to give me a one-sentence answer on this. Are you ready?

[00:26:28] William McKnight: I’ll try. Yes. One sentence.

[00:26:29] Scott Kinka: Let’s do our best. All right. What do most tech leaders fundamentally misunderstand about what it takes to build a real data strategy?

[00:26:38] William McKnight: Data strategy is a thing that holds its own in the organization and supports many applications. Data’s not aligned to an application, or shouldn’t be thought about as aligned to an application. It’s aligned to an enterprise.

[00:26:54] Scott Kinka: Got it. Got it. You said earlier, I love this, data is not a drag along. Explain that concept.

[00:27:02] William McKnight: Yeah. Well, so many times you’re building an application. An application can be 50% or more done with data. You just wanna drop a thin application layer on top of great data to accomplish the goals of the application. You can’t do that if your data’s all over the place. So when I say what I said there, what I mean is that we want specific data skills to manage that data and bring it up to standard, not someone who’s an application developer that does data on the side.

[00:27:37] Scott Kinka: What is the biggest mistake you think companies make when they’re treating data management like an IT project rather than, you know, part of their overall business transformation?

[00:27:47] William McKnight: They’re missing out on so much when you’re still treating IT like a cost center or something like that because, really, the ideas that are going to take businesses forward today are coming from the IT team, coming from the technology team. And if you shut them off, you’re shutting off so many great ideas that are gonna take the company into the new century. So really don’t do that. Give them a voice at every table because that is where the great ideas are coming from.

[00:28:23] Scott Kinka: I love it. In ten years, we’re gonna look back at this moment in enterprise AI and all the underlying things that fed into it, including data strategy that we’ve been talking about today. What will we all look back on and go, man, I wish we had done that differently or paid more attention to this. What do you think we’re all gonna look back at in ten years and think differently about?

[00:28:42] William McKnight: I think that it might be something similar to how we’re looking back today, ten years, and saying, why didn’t we take care of that back then? It’s just data. It’s just making sure that we attend to data as the primary asset of the business that it is, and we quit just duplicating data all over the place and building data to different standards and creating all this technical debt around data. So have a great architecture. That’s what we’ll regret if we don’t attend to it now: having a great data architecture. Once you do that, like I said before, you can drop any kind of thin application layer on top of great data, and away you go. You just cut the development effort by 50%.

[00:29:27] Scott Kinka: Yeah. I mean, so said in a different way, data is your intellectual property, not an artifact of what you do every day.

[00:29:36] William McKnight: Yeah. I like that. Absolutely.

[00:29:38] Scott Kinka: 100%. Okay. Give me a contrarian opinion in your market around data strategy that you feel like listeners who are data nerds might push back on? Something where you’re controversial.

[00:29:51] William McKnight: You know, we do the, like I said before, we do these benchmarks, right? And people think that all databases are equal in terms of their performance. And if not, it’s probably a flawed benchmark or a gamed benchmark or something like this. No. Not necessarily. Not at all. These products have variations in how they are architected, which will manifest in their performance as you start to throw different types of queries against it, be it analytical, operational, AI, anything in between. And so it’s really important what tools you select. It’s not just, oh, I’ll just pick one of many, or I’ll pick whatever’s in the upper right of Gartner or what Joe over here said or whatever. It’s important. It’s important enough to really pay attention to, and sometimes that means stepping outside of the enterprise contract that you may have with one of these big players. It may even actually be worth it to get a best-of-breed database, data integration tool, what have you, even if you have an enterprise contract with one of these major players. Well, it’s interesting.

[00:31:07] Scott Kinka: You’ve done some research. I’m kind of dismounting for a moment from the quick fire. Um, you’ve done some research that found companies were built, you know, companies that build a fragmented multi-vendor stack are paying something like three times more over multiple years than those with unified architectures. I’m probably screwing that up, and I’m sure you know that what I’m talking about is from your research. Can you unpack that a little bit? Is that what you mean? Like, it might be worth taking a step back because you’re gonna spend the money eventually to just get it right as single stack as possible to deliver the solution.

[00:31:44] William McKnight: Yeah. Yeah, that’s right. And I know the research you speak of, and that is roughly the conclusion that we did come to. We actually engineered an agentic environment, um, in a POC approach, right, but we actually engineered it a couple of different ways: the do-it-yourself approach of cobbling together tools out there or starting with a foundation. In that case, I know it was SAP. Starting with the foundation of tool sets that had been largely integrated gives you that push forward. So, not to be contrarian to what I said earlier because there’s a time and place for best of breed, for sure, but you really gotta know your stuff today and know what you’re getting into because the tools you select, the platform you select, whether you go with fully integrated like this study suggests or otherwise, it’s going to drive your TCO by a good three- to five-fold over the course of time, and you may not even know it for years. So you talk about regrets. There may be some regrets that we didn’t pay more attention to the tools that we picked to do the job with down the road.

[00:33:04] Scott Kinka: 100%. No. That makes a lot of sense. Um, two, three quick ones. One, give me a bold prediction of any kind. It can be in tech. It could be in sports. It doesn’t matter to me. Just give me something we can look back on in eighteen months and see how right you are.

[00:33:21] William McKnight: Oh, the Spurs are coming back. No, I actually, I don’t know about that.

[00:33:26] Scott Kinka: But mind you, by the time this episode airs, people won’t realize that the night before this taping was actually when the Knicks came back at MSG after being down 27 points at halftime. So all right. That’s not your prediction? You’re coming off of that? It’s not the Spurs coming back?

[00:33:42] William McKnight: My prediction, and um, buckle up for this one because it’s not pleasant to talk about, but the more prepared we are, the better off we’re going to be in these life circumstances that we’re going to find ourselves in very soon. I don’t know, five to ten years, and it’s gonna last a good while until we learn that AI can actually improve our lives, and we start to treat it like that. I don’t know that we’re treating it like that right now. I think we’re treating it like it’s a profit maker. Okay. Fine. That’s how we get the ball rolling. But as that starts to come into the employment arena and we do start to see some real challenges around employment because of AI, we’re not equipped to deal with that because we have this ethos of we gotta work hard. This is America, you know, where we are right now anyway. This is America. We gotta work hard, and if you don’t, that’s on you. That’s on you. And I don’t see that abating anytime soon, but eventually, it will have to because things will get worse in the employment arena. So be prepared. Don’t be doing jobs that can be easily automated because chances are it will be.

[00:35:04] Scott Kinka: Yeah. It’s a really good point. Um, it’s funny too. I love what you said about, you know, this is America. I mean, I’ve said on this show a couple times, we still have this, like, World War II, we-can-do-it work ethos where, like, the number of hours you put in is somehow directly equated to your productivity regardless of how many widgets you ship. Right? Or, you know, how much work you put in, I should say. Like, I’m just thinking about it from a tech arena. You know? Like, how do you judge the person who stayed in the office for ten hours to produce the same amount of code, you know, as somebody who stayed home and produced it in two hours with AI? You know? I we do not have the HR acumen yet to be able to evaluate that and appropriately hire, fire, bonus, etcetera. I am equally worried in that bucket. Super interesting. All right. Two, what’s on your nightstand? Are you reading anything interesting that you’d recommend?

[00:36:06] William McKnight: I read some science fiction, some I don’t know, what am I reading now? Some Asimov, something like that. That’s obviously not science fiction strictly speaking, but I read on subjects like that. I read a lot of books in our area of data, books from colleagues, books I’ve been asked to review, and things like that. I find that over time, though, that I’m more drawn to short forms of books. Just give it to me. Give me the bottom line. And I just read the book summaries. That’s primarily what I read now: book summaries, and I can bang them out, you know, one a night. So you can get main points. So I do a lot of that. Yeah.

[00:36:49] Scott Kinka: Okay.

[00:36:49] William McKnight: Primarily read now is book summaries, and I can bang them out, you know, one a night. So, uh, you can get main points. So I do a lot of that. Yeah.

[00:36:59] Scott Kinka: I love it. It’s so funny to go back and read Asimov right now with everything that’s going on. Most of this conversation, it’s just almost terrifying how, you know, pertinent it is to read today, in a lot of ways. I recommend it to a lot of people. Last one. Here’s the fun one. Where since you’re a sci-fi guy, some dystopian future occurs. You can choose whichever one you like. Um, could be the aliens, the zombies, the end of the world, you know, you name it, you tell me. Um, but we’re in that. And you have one app that still works on your phone. Which one is it?

[00:37:35] William McKnight: One app that still works on my phone. I, if doom is around the corner, I mean, I might want something that talks, something just to talk to me, and what’s better than an LLM?

[00:37:53] Scott Kinka: So your chat agent is the answer. You’re just gonna need someone to talk to so you’re not fully alone. Um, I love it. This was super fun. Um, for listeners that wanna follow up on your work, maybe read some of these benchmarks, engage with the McKnight Consulting Group, what’s the best way to connect with you?

[00:38:11] William McKnight: Through the website, which is cgforconsultinggroup.com. McKnight CG.

[00:38:20] Scott Kinka: I love it. Um, well, any last words of wisdom based on the conversation that we had today for the CIOs and IT directors who are listening and trying to figure out what they should be doing around data?

[00:38:31] William McKnight: I mean, I just find that where you are is important, and that’s one thing, but the momentum that you create is more important. So if you’re standing still, you’re actually going backwards in data. You gotta be moving it forward. If you’re still doing everything the same as you did a year ago, even five years ago, then you are way falling behind today. So you’ve gotta pay attention to it and get your momentum going in the right direction. And once you do that, you gotta attend to it, keep the ball rolling, but you can fulfill your dreams with data.

[00:39:13] Scott Kinka: Get that momentum going. I’m fired up. I’m gonna run. I’m gonna go run through my door here over here. You got me all fired up. William, I love it. Thank you so much for the time, and we’re looking forward maybe to talking again at some point. This is a wrap on this episode of The Bridgecast. If today’s conversation gave you something to think about, please hit subscribe, leave us a review, share it with your team internally, some great nuggets for your data teams internally to think about on this episode, and we’d love to see you and them on a future episode of The Bridgecast. Thanks for listening. Listen. If you made it this far, there must have been some good content, or you’re just a fan or a relative of mine. In either case, we appreciate you spending your most valuable asset with us, your time. We don’t take that lightly. We’d also like to say one more time that we’re appreciative of Bridgepointe Technologies and their belief and sponsorship of this show. We hope that if you or someone you know is thinking about your company’s digital transformation or simply the next IT project that you may not have the resources, budget, or time to get to, I’m sure that Bridgepointe, one of the country’s fastest-growing technology advisory and procurement firms, can help. Check out bridgepointetechnologies.com. Don’t forget the e on Bridgepointe, or simply reach out to me at skinka, s k i n k a, @bpt3.net. Also, take a minute, please, if you would, to give us a five-star review on your favorite platform. It helps give us the visibility to reach other people like you. Thanks for listening to this episode of The Bridge. I’m Scott Kinka, and until next time, there’s a lot of noise out there in business and in life. Do what you can to be the signal.