 Hello and welcome. My name is Shannon Kemp and I'm the Executive Editor of Data Diversity. We'd like to thank you for joining today's Data Diversity Webinar, Data Governance Strategies. The most recent installment in the 2015 monthly Webinar series called Data Ed Online with Dr. Peter Akin, brought to you in partnership with Data Blueprint. Now let me get the floor to Megan Jacobs, the Webinar Organizer from Data Blueprint, to introduce today's speaker and today's webinar. Megan, hello and welcome. Thanks, Shannon. Hello, everyone, and welcome. Thank you for finding the time to join us for today's Webinar on Data Governance Strategies. As always, a big thank you goes out to Shannon Data Diversity for hosting us. We'll get started in just a few moments after I let you know about some housekeeping items and introduce your presenter. We have one hour for the presentation, followed by 30 minutes of Q&A. We'll try to answer as many questions as time allows, but feel free to submit questions as they come up throughout the session. To answer the top two most commonly asked questions, yes, you will receive an email, links to download today's materials in the Webinar recording so you can view it afterwards. These materials will be sent out within the next two business days. You can find us on Twitter, Facebook and LinkedIn. We've set up the hashtag Data Ed on Twitter, so if you're logged on, feel free to use it in your tweets and submit your questions and comments that way. We'll keep an eye on the Twitter feed and we'll include answers to those questions in our post-session email. Now, let me introduce you to our presenter. Peter Akin is an internationally recognized data management thought leader. Many of you have already know him, or have seen him at conferences worldwide. He has more than 30 years of experience and has received many awards for his outstanding contributions to the profession. Peter is also the founding director of Data Blueprint. He has written dozens of articles and eight books. The most recent is Monetizing Data Management. Peter's experience is more than 500 data management practices in 20 countries and consistently named as a top data management expert. He has spent multi-year immersions with groups as diverse as the Youth Department Defense, Deutsche Bank, Nokia, Wells Fargo, and the Commonwealth of Virginia. Now, it appears at conferences and is constantly traveling. So, Peter, where are you today? Today I'm in Richmond and I'm getting ready to head up to DC for some work up there, but it's good to be back and really we were thoroughly excited because Shannon and the Enterprise Data World Conference sold out the week before last. So, we had an absolutely packed group of people up there. I don't think I've ever felt that kind of excitement around really what we believe is our core mission here. So, and just to put out this as well, we're doing data governance today in hopes that we'll see some of you all at the Data Governance Conference that's coming up in June. So, again, looking forward to carrying on the discussions that we started last week on all of these things here. So, really collectively, we all gathered in DC and sort of recite this credo that data is our sole non-depletable, non-degrading, durable, strategic asset. Many people have been talking about data as the new oil, but oil is kind of icky. So, then we saw somebody put up this thing that said data is the new soil and that's kind of cute. I have things in it and stuff grows. That's really kind of cool as well. We actually did find this T-shirt. Data is the new bacon. Yes, okay, great. Whatever it takes to get people to pay attention to this stuff. Because if we can help everybody collectively unlock business value by strengthening your organizational data management capabilities, providing tailored solutions, and building lasting partnerships, we will have achieved what we set out to do. Now, today's specific focus, of course, is subtitled if you don't know where you're going, any road will get you there. And many organizations have fallen into this trap most famously, of course, is Target. Beth Jacobs was fired, actually she resigned, the Target CEO was fired. But what was really interesting and got a lot of attention is that last summer a group called the International Shareholder Services, which represents big shareholders on how to vote corporate issues, said that really the blame should go all the way to the board of directors of Target. This stuff hits the board of directors agenda. You know that this stuff is getting on people's radars. So Target had their issues. We're not going to talk a whole lot about them, but it is important to learn how to work with these. And Home Depot, even though it happened just a couple of months after the Target breach, their CEO had not been prepared to talk about this. So after the breach was discovered, he put out a statement saying Home Depot is working around the clock to find the data breach. That's not something you want to say if you haven't actually found the data breach. So they had found it. Number two is off message there. And it was a real challenge. So we're going to start off our webinar today talking about strategy. Look at relatively recent usage. It's been a term that's come online more recently, although people really do understand it today. And it's going to present a series of difficult choices for us, because we do have to understand that governance strategies are going to be contextualized. And I'll give you a good explanation on all of that. They won't dive in and see what really is data governance. Why is it important? And more importantly, why does it need to be governed by a strategy? We'll look down at some of the basic things that you need to have as data governance. And one of them will be frameworks and others, a series of building blocks, checklists, and then things to avoid all the way around. I'm going to give you a couple examples of governance and action by storytelling in particular. If you want your governance program to be successful, you do need to develop some storytelling abilities in order to do this. So let's jump in. Strategy, as I mentioned before, really in the 1950s, if you had used the word strategy, people would not have known what you meant unless you were in some sort of a military context, because most people just did not use the word. The good definition for it is a system for finding formulating and developing a doctrine that will ensure long-term success if followed faithfully. A great definition. A little bit wordy there, the one I like a little bit better is a pattern in a stream of decisions. Notice neither of these go to a three-ring binder that gets placed on a shelf somewhere. It's more a way of being and living something than it is specifically looking at a doctrine, per se. So let's take an idea around strategy and one of the ones that some of you may be familiar with is how do I defeat the competition when their forces are bigger than mine? I remember our definition of pattern in a stream of decisions and the answer is divide and conquer. So I'm going to show you a little bit of an animation here and the idea is not that we should have necessarily a history lesson, but that at one point in time, Napoleon was trying to figure out how to beat up the other two armies. Now Napoleon's in blue. The other two armies are in red and black. And his answer was if I hit them really hard at their intersection where the two armies were joined together, they would retreat in the direction of their supply lines. And then I could turn my attention to first one and then the other in order to defeat them. And of course this is what happened. He hit them very hard. Then his troops were agile enough to turn around and defeat the Prussian shown in black and then of course the British in red. Now strategy, pattern of decisions. His focus here was first of all hit them hard and make them move back. Don't gracefully engage them, but really do a pummeling in this case. And then go after one and then the other. So his troops knew they would be in for a longer than normal action. Again, it's these patterns in a stream of decisions. Sometimes people don't like the military pieces and sports analogies go very, very well also. So we look at Wayne Gretzky who's been quoted many, many times as saying that his strategy, the thing that made him a great hockey player is not that he chased the puck around the floor, but he chased, excuse me, he went to where the puck would be. And when he was able to put himself in the position to score goals on a repeated basis, it made him into a great player. Now these are things that you're going to have to know. And as I'm working through this, I'd love to tell this particular story that blue front often we get calls for people, as you might imagine, to come help them develop a governance strategy. And so one of the first questions is we say, what is your organizational strategy? And if they can tell us, that's great. If they don't, we have trouble with that because it's very difficult to put this in a non-contextual form. At least if you have an IT strategy that is derived from your organizational strategy, you may be able to articulate something that would allow you to go forward there. Frankly, the IT strategy is less important for us than the organizational strategy because data is a business asset. It's not typically an IT asset. And so we want very much, and if we have a choice to derive our data governance strategy based on the organizational strategy instead of starting with an IT strategy specifically. Now many people are familiar with the terms governance and corporate governance is just considered to be good form to have people who are responsible for certain aspects of the way things are done. So in the olden days we wouldn't have necessarily talked about corporate governance. Now we get into IT governance as well. And you'll see here again you've got a combination of computer auditing, information security management, risk management. These are all areas that are becoming more and more understood organizationally. Yet when we go back in and look, and I have to give a shout out here to my colleague John Ladly who put this wonderful illustration together a couple of years ago to show that many times in many organizations there is a gap between the business priorities and what's going on in the IT initiatives. And the reason for that is because it's very difficult for many organizations, particularly large organizations to start off with an IT strategy and actually make it work, excuse me, an organizational strategy and make it work at the IT strategy level. And that's because IT is implemented according to a series of projects. So in this hypothetical organization here, you see on the left-hand side of your screen the organizational strategy, and that gives to a specific set of goals and objectives and eventually winds its way into IT. But by the time it gets down to the IT project level it is very often confused and not well understood. If it exists at all it's likely to not be a singular set of strategy and objectives, but that this is perceived at the project level as really project-specific goals that are often confused, inaccurate, and incomplete. And the bottom line is that very often IT projects do not well reflect organizational strategy. So again, our preference from data governance perspective is to move your data governance strategy to be something that is derived from your business strategy as opposed to a specific IT strategy. Now let's look at some hard choices that you have to make initially. And many people are unaware of these, but they go back to the basics of, in particular, organizational strategy, which is derived largely from Michael Porter. Many organizations fall into the first quadrant where they're really not trying to do anything too innovative or much at all with their data governance components on here. Some organizations, however, end up being focused on efficiency and effectiveness and a company that comes to mind, as you might imagine, is Walmart, who are experts at making organizational effectiveness efficiency a part of the culture of their organization. And you all know this because their strategy is every day low cost, which translates to the customers as every day low price. A very simple, very clear example of how to do that. Another company that you might think of as focused on innovation then is Apple Computer, where they use data to create strategic opportunities. Now, the question that has to occur to most people here is, are the people who are really good at the things in quadrant 2, increasing organizational effectiveness and efficiencies, also the same people? Do they have the same knowledge, skills, and abilities that people are using data to create strategic opportunities? And the answer is typically not. Similarly, the people in quadrant 3 also are not typically focused on effectiveness and efficiency. So you can't really do both. And you do have to say down and say, what should we be focusing on first? Now, it may be perfectly reasonable to say, first, what we should do is efficiency and effectiveness and then create innovation or the other way around. But when we look at these things over time, we just don't find most organizations, in fact, doing either of them formally. And that's why it's so important to try to put in place something like this. Now, this is a supplemental diagram. I'm not going to walk through it here, but the CMMI Institute has put out some guidance on this. And Melanie Mecca and I gave a talk two weeks ago in D.C. on this. And we're going to carry that talk on at the Data Governors Conference coming up, talking specifically about data strategy elements, focused in this case on quality in there. So I'll just put that up for your reference. We'll move now into the next component here, which is really looking at definitions. And any time that you're out on the Internet and you start to see lots of definitions out here, and these are all very good definitions, but it doesn't unfortunately really help us to understand what's going on from a data governance perspective. So I'll just read a couple of them here. Formal orchestration of people, processes, and technology to enable organization to leverage data as an enterprise asset. Absolutely. Convergence of data quality, data management, business process management, risk management surrounding the handling of data in an organization, great Gwen's definition from the Data Governors Institute, system of decision rights and accountabilities, very clear definition there. For information-related processes executed according to agreed upon models, which describe who can take what actions, with what information, and when under what circumstances using what methods. Again, very, very good definitions here. Rob Siners from KIK Consulting, he's the editor behind TDAN, which I understand is getting a reboot sometime in the next quarter here. If you haven't had a chance to go to TDAN.com, again, very, very good publication that Rob's been working on for many years out there. Anyway, his definition, the execution and enforcement of authority over the management of data assets and the performance of data functions. All of these are good definitions. I won't continue to read them here, but it really does become a challenge for us. Now the last one here comes from the DM BOC. And if you're not familiar with the DM BOC, the DM BOC is DAMA International's definition of the body of knowledge. And we've started now to work on these, and we're hoping that through these seminars and other things that we do, many of you will be, in fact, able to qualify for our designation called the Certified Data Management Professional. I won't dive into the details. There's lots of information on the web, and we had several hundred people take exams at the conference on this. Another reason for going to these things. Really, when you think about data governance, what you're trying to do is make sure that data is accessible, it's secure, it's consistent, it's high quality, and it's auditable. In other words, where did the data come from? Is the lineage clear? Does IT know who's using it, and for what purpose? All great focuses in here. So my comment on all of this is that data governance should really have a purpose statement, which is to say, what does data governance mean to my organization? It's fine for those other definitions. They're great. Actually, I like this definition. It's managing data with guidance. Now, managing data with guidance means that what you're doing is taking a resource that was previously not as focused, not as well managed, and putting some guidance around that process so that people are, in fact, able to see what's happening. It means for your organization, particularly those of you that are getting started, or let me also suggest rebooting your data governance organization because we've seen an awful lot of people that sort of took their first attempt at this and didn't have the results that they wanted to get, and now they're looking at a rebooting of data governance on this. So it's getting some individuals whose opinions matter to form a body, and the body needs to have a formalized purpose, some sort of authority, who can then go out and evangelize, advocate, but not dictate, enforce, or rule with a process that we describe as data-centric development practices, increasing the scope and rigor of those practices. Because if we can do this, people will see that these things become relative and relevant. Now, many organizations, and particularly when you think about upper-level management, they have a lot of things on their plate, so they're going to say, why should I pay attention to this to the exclusion of something else that may be of interest to them? And the idea is, you know, yes, this stuff is important, and so are many of the other things that they deal with. It's absolutely critical to talk to them in ways that make sense to them. So we always help the organizations find, if you will, their own voice relative to data management. And this is one of our groups that we worked with where they literally said getting access to data around here is like that Catherine Zeta Jones scene where she's having to get through all of those lasers. Again, some of you may remember this movie. There's all these lasers, and she's trying to achieve a jewel theft in this process. And she keeps working on it, but this spoke to them. This meant something to them, and when we showed this little video clip and they started to use language around this, management kind of got the idea that, wow, if they're having to do that much effort in order to actually get access to data, which is the thing we want them, of course, to do as knowledge workers, there's probably something we ought to pay attention to here. Now, there may be other types of articulations around this, and I'm showing in the center of this diagram one of the pieces that Melanie and I will be focusing on will be focusing on the green, the data quality aspect of this, but this is what the new data management maturity model focuses on, and of course, you can see how that's going to be important, and it may involve decision-making. Just try to decide certain aspects of what's going on, or it may involve data quality challenges. It may involve performance, again, the efficiency and effectiveness things we talked to a little while ago. It may talk about data security, and my colleague Lauren Joseph and I presented at the Commonwealth of Virginia's Information Security Conference an integrated approach to looking at governance risk and compliance at the same time as you're doing data governance here. It's a very, very nice integrated platform. Most people are not looking at that. We find out that they're actually looking at the same thing twice, which is, of course, less efficient, or even going back and using data from a strategic perspective, as we talked about before, the innovation component that we look at. So let's take a quick difference here between data governance and data management, because many people like to sort of be able to have an elevator speech around this. Governance is high-level policy guidance. It's general directions and guidelines. And you might say, for our organization, all information, not marked public, should be considered confidential. That would be an example of some data governance that's in there. It may not be right for your organization. Don't take that away as what it is. That's what it should be. And that's the implementation of this, the function of planning, controlling, delivering assets, and delivering data in support of specific challenges. Now, I mentioned before this DMM structure from SEI. We're very, very pleased that this is out and able to be used as a standard against most other things. Again, there are different talks we'll have on this later on in the year, and if you don't get a chance to see Melinda and I do it together, I'll share it as many conferences as I am this year. But each of these practice areas needs to be governed in a way that makes sense. And the easiest way to describe these is with something called the capability maturity model, which is what SEI has put together now for data. At level one, everything is informal and depends on heroic efforts. At level two, it's managed. At level three, it's defined. At level four, it's measured. And at level five, it's optimized. It's a very simple model, and more importantly, most managers have become exposed to this due to the high adoption rates that this has occurred throughout the last 10 years of this. This is the basis for TQM, TQDM, ISO 9000, all sorts of other process improvement frameworks that are on here. And when I take these two pieces, the five model that I showed you before on the left-hand side here of the stair steps, I can now rate each of these areas. So one of the first things to do as an organization that deciding that it needs to contain more governance around their data is to do an assessment in that area and see where they look at. And here's what I did for the insurance industry a couple of years ago, where we look at these five areas and then simply say, where are they? And the average insurance company that was surveyed under these circumstances did not actually have managed data government practices. What it meant was that their data governance occurred at the work group level and that meant that there was room for many, many varieties and very little standardization that occurred in here. So it's a very good chart to be able to show people. Here's another one that I did for an airline that was looking at this. Again, if you look at the answers here out of context this might seem okay, I've got not really sure what that means and I say, well, here's your competition and all of a sudden I've got everybody in the room's attention because the executives do understand that they're the ones and their competition was a two. Notice I'm remaining keeping the airline anonymous in there. Then of course we want to take the ones and make them into twos which says this is really what the focus of the data governance program is going to be about for the first phase of its operation. We're going to try to take those ones and turn them into twos. There's three ones on that chart. We'd like to have them all be at level two by a certain data and that becomes the focus of data governance initiatives. Now don't feel badly if you're looking at this and saying, well, gosh I'm not sure where I am. This is what I call the era of big data. Notice it ended in about 2012. Different topic we can do on that are relatively unchanged over the period of time that we looked at all of these levels and that's generally not good because when you consider how much data is in fact coming out these were the numbers for the London Summer Olympics game and these are big numbers. I like particularly the last one there were more devices connected to the London Olympics than there were people on the planet. That's a pretty good statistic there. What are you trying to do with your data governance? The idea is you're trying to focus around policy standards architecture procedures and metrics. You're trying to look at regulatory compliance and conformance to policies. You're trying to prove the way data is delivered to projects and services and you can't do that if you can't track them and oversee them. You of course want to resolve data related issues and lastly promote the value of data assets. The activities then that you're going to be doing in here are to understand the enterprise data needs to develop and maintain a strategy around that to establish professional roles to appoint stewards. Now I'm going to make a point here on the stewards in particular where I saw one organization gave me their data governance plan and because it was a quarter into we're going to appoint the first round of data stewards, excuse me, we're going to appoint all of our data stewards by February 1st. I said, oh, stop for a minute there. Probably, will you get it perfect the first time you appoint your stewards? They said, we're good, but we might make a mistake or two. I said, how about instead of appointing all of your data stewards by February 1st, how about you appoint the first round of data stewards by February 1st? We're going to do a second round and if we miss somebody the first time we may be able to catch them the second time or mistakes we may be able to improve the process. In other words, what are the likelihoods we get anything perfect the first time as human beings? This gives you a little bit of time around that and of course the other part of it was don't do it on April 1st, it's just the wrong day. Let's do something else. Look at an architectural component plan and start to sponsor projects and services that are around data management and help to start estimating data asset values and costs. You'll see in a little bit we'll get to a place where we've described how much goes into the process of data related issues in IT. Our measurements show that between 20 and 40% of all IT costs are spent resolving data issues that can be handled better by governance policies. When you start to look at the IT bills many organizations are faced with, they then also start to pay a lot more attention to very specific deliverables in there that allow them to now understand and move forward in better development practices. The primary deliverables from your governance organization are going to do things like late-weight data policies, a first couple of data standards, a count of resolved issues and an impact analysis of those issues. Looking specifically at funding and making sure that they are successful data management projects and services, quality improvements over time. Many organizations unfortunately can say that the following statement applies to them. All of our data quality, excuse me, all of our data is of unknown quality. Well, that's a little bit scary and I get organizations to say that not publicly obviously, but to certain key individuals and people go, oh, I don't like that. All of our data is of unknown quality. That's not good, is it? The answer is that's not good. And finally starting to recognize the importance of that data. Again, several things in the news recently that will cause people to pay attention to this. It also means that your governance roles and responsibilities now have to include different types of stewards. I've listed three types here. They may be appropriate for your organization. They may not executive stewards or some power. They have the gravitas, if you will, that's going to pull into place. These things, the coordinating data stewards are people maybe who are either executives or do their business. Maybe they're working with IT. Business data stewards coming obviously primarily from the business side. Data professionals, some sort of data management executive or as I like to call it, the top data job. Working in conjunction with the chief information officer or such. There are the suppliers of the data pieces again, the business executives, IT executives, stewards, regulatory bodies and things that we're going to focus on in the external environment, ambient data sources. And consumers, of course, the knowledge workers in our organization are going to be focused in these areas. So again, the goal is not to sit down and put all this in a plan that is going to work perfectly from step one to take some small steps and crawl, walk and run your way up to a more professional program. Similarly, the tools that you will need in order to do this are very, very easy to do. If you've got an email, if you've got the workflow that you can throw at it, that's great. Many people think that in order to have data governance, you must have a metadata repository. And the answer is actually that you must have something that functions like a metadata repository. One of the best things that a data governance organization can do, and I've seen this happen many times in the past 15 years, is that they have very good products that work very, very well. But it's a fairly sophisticated product being sold to an organization that is perhaps not as mature or not ready to handle that. It's kind of like handing the keys to the Ferrari to a 16-year-old who's just gotten their license. Yes, absolutely. They know how to drive, but there's a responsible way of doing it. So what I like to tell organizations is don't go out and buy a repository. Make one of your own in a relatively low-cost fashion. You can build a metadata repository in a week. I've got a webinar out there that describes it in another part of the website here that you can get to and find out about it. And after you play with that toy webinar, excuse me, that toy repository that you put in place for a year, then you'll be ready to have a mature conversation with the vendors. But until then, there's such a mismatch between the two, it really does not result in a lot of adoption of these things. And my colleague Dave Eddy did a study on these things and found out the half-life of these repositories was about five years. Just not good as well. You can use your existing issue management ticketing systems to work on these issues. Sometimes it's just a matter of adding a couple of metadata categories to that. It doesn't hurt to end up with some governance so that you have some, excuse me, governance dashboards so that you have the ability to track these things and to see what's happening overall because after all, management is eventually going to come back and say, oh, you've been governing data for a year, what did you achieve? Well, what we'd like to see is that you achieve some value of the data that you understand a bit more about how much it costs. Are you achieving objectives that you've set out to achieve? Are you getting some decisions to what sort of steward coverage do you have in the nationals headcount do you have? Again, these are things that you can come up with which will eventually lead you to the process of assessing your overall data maturity in those areas. Again, these things cost millions each year in terms of productivity, redundant and siloed efforts, poorly thought out hardware and software purposes. Some organizations are able to shift from reactive to proactive. Many organizations delay their decision and there's my statistics, 40% of IT spending can be reduced through data governance. Now, again, as I look at these things over large numbers of industries, hundreds of samples, I find that 70% of them return as close to zero as you can get. There's little or no actual value that comes out of it and only 30% of them have any positive return and that about 11% actually achieve positive return on investments. So there's two pieces to this. One, first of all, if you do this, the competition at this level once you start to make significant contributions becomes very, very easy to quantify. And secondly, once you've started to do this in a way that is tangible, you can get better at it. And of course, that's really the key, we don't want to just do this and do it at it the first time through, but the second time, third time, we will actually get much, much better. The reason this has been problematic is because we've had some terrible, terrible guidance that has gone to our people that have come through and learned how to do this. This is because we do something called application-centric development, which means that an organization starts out with strategy and moves to goals and then, reasonably enough, drive the development of specific systems as they're going through this. There's a problem with that. We'll get to it in just a second. Because once we start talking about systems, then, of course, we need a network infrastructure to put it in place and that data becomes an afterthought in this entire process. And the problems with this approach are that it makes sure that the organization-wide information requirements that any process architecture is narrowly formed around applications and that very little data reuse is possible under the circumstances. Now, when we measure what we're looking at across things, the amount of data, excuse me, the amount of code reuse is very, very small. Again, developing systems, as soon as we say something at step three that system X is going to be purchased, it sucks the air out of the rest of the conversation and we are not able to go through. The reason is because we haven't taught people for years that information is an asset. It's an economic resource. But if you don't own and control it and use it to produce value, it's very difficult to justify the care and feeding of the data or the ability to put data to work in very, very significant ways. Now, the fiction that I described to you earlier is that systems produce the development of data requirements. But if we think about it, our ability to create information systems components is a creation-oriented activity. When we are developing system components and capabilities, we are taking a capability that didn't exist and adding that capability to the organizational repertoire. Data is different. Data does not follow that pattern. Data evolves. That goes from one version to another. And in that process of evolving, you will also note that it doesn't begin and end as all IT projects should. So the idea here is the first thing that your data governance initiative should be doing is raising awareness that we happen to have taught everybody incorrectly that they can develop their data and their system requirements concurrently. The only thing that can happen given those requirements that are developed concurrently is more small piles of data. Now, that's generally what people are trying to avoid. So the way to fix that is to make sure that everybody in the organization understands that data evolution is separate from. It is external to and that it must proceed system development activities. If you don't understand your data needs, you shouldn't be developing any software and you certainly shouldn't be buying any software. This simple change has a profound effect because if we now look at what we consider to be data-centric development practices, we start off with exactly the same changes that we're going to see. We start off with the same two first steps. An organization has a strategy and they have some goals and objectives. In this data and information, there becomes the next thing that they should develop. In order to achieve our strategy, we are going to need the following oil, soil, bacon, whatever it is that you want to call it. This is where the information layer needs to be developed. Then the network infrastructure comes in and finally at the very end of that process, we now start talking about systems. By delaying the system's implementation until a later stage in the process, you now have the ability to develop information or data assets from an organization-wide perspective. You're now able to support organizational needs in a way that complements existing process flows and finally you are maximally able to reuse data and information instead of minimally reusing software, which is where most organizations are when they measure them. Again, from a data-centric development perspective, it's an architectural practice, it's an extension of existing capabilities. There are some challenges that are unrecognizing your organization, which means your technical engineers have different skills that they need to have. You're going to have to be a governance organization devoting to communicating these ideas. There's an absolute scarcity of professionals in this area and it does speak to the needs of a specialized discipline. There's a quote from the book, obviously, monetizing data management up there. When our organizations begin to transform to a data-centric approach, we begin to measure success differently than we did before. The same project, the data management process, valuing the correct data more than on-time and within budget, valuing more correct data, more than correct processes and auditing data rather than auditing project documents. Linda Buvalow contributed a chapter there where she was able to, in fact, describe a very, very effective cost-benefit analysis in that particular process. I won't ruin her story for you, but if you need to, if you need to. So these requirements, then, are understanding that data is very, very different from the way it has been. And in order to get started on this, you have the standard startup process. The left-hand side says assess our context, that's the assessment process that I described to you earlier, define a data governance roadmap, which is not to say plan out every aspect of how data governance is going to work in what are we going to do first, maybe second, and possibly third. Secure some executive mandates and make sure they understand that they're not going to be doing this in a complete plan, that it's going to be an evolutionary plan and then assign yourself some data stewards to get things started. Execute your plan, evaluate the results, revise the plan, apply change management and repeat, just like shampooing. Lather, rinse and repeat. And plan to get better at it as you go forward. Now some of the essential ideas that are in here have been contributed by our community to this generalized collection here of data governance frameworks. I'm going to flash through a bunch of these quickly for you. But the frameworks are a system of ideas for gathering, for guiding analysis and organizing the data around these projects. They're all about making decision-making priorities so that you know which decisions are going to be first. Again, in this framework here, the next step clearly is to put the roof on the house because then we can work inside the house as opposed to putting the floor in before we put the roof in. It does give you a means of assessing the process, so again putting up the walls first, et cetera, et cetera, to make it all dependent on what you're looking at here is an ipo model, input process and output. Again, process is represented by the activities in the teal, the tools that we talked about down below on the top, the goals up at the top on there. Again, I'm not going to walk you through this chart. What you should do, however, though, is look at a series of these charts. Here's the second one here. Oh, and by the way, that center part is the part that actually does apply to the CMMI piece we talked about a while ago. Again, from Gwen's data governance institute here, 10 components of the data governance program, these are neither right or wrong. They are good guidance on how to go about thinking about this process. Here is Bob Siners mentioned his definition before. None of these are right or wrong. The question is what makes sense for you? So by looking at these several different models that we have here, you can now look and say this work for us. So here's one from IBM. Here's a second one from IBM that they have. So again, just two different ways of looking at the aspects of data governance. Here's one from SAS Institute, Jill Duchay's organization baseline consulting, talking specifically about it here. An interesting one here, the American College Personnel Association and they've made theirs look like a ship. Okay. Not going to be right or wrong. This is something obviously that works for them. The point here is to look at all of these and see what works for you. So NACIO, another organization that we'll be speaking with coming up in the near future here also has a process for putting these in place. I show you these not because any one of them is right or wrong, but in fact to give you ideas to compare and contrast. And what would make sense really if you're just starting out and or rebooting and I really am seeing many organizations that are going through a data governance reboot. What they do is they try them on. They sit down and talk about them. What are the advantages and disadvantages? And usually by going through those six or seven of them that you saw there, you will find some aspects that start to make sense for you and then you can make something that says what does data governance mean? If we look at this in forms of a checklist, what we're seeing here then is do we have authority? Do we have some policies and procedures that we're starting to work on? Have we started the process of inventorying our data? And by the way, I'll give you a quick stat on that one as well. 80% of your data out there is rot. It's redundant to obsolete or trivial. So you probably shouldn't be wasting any time or time. So what we're seeing here is that data governance is defining the criteria for useful data and getting rid of the stuff that you found. Like having your house go through spring cleaning if you have never cleaned your house in 2000 years. Your house is going to be pretty damn dirty at that point. Again, starting to look at records management, quality data access and security and risk frameworks in place. Again, these are just things that you can put in place. It's a very, very nice five-page paper that's got a very good set of guidance in order to do this. Here's another scorecard from back at Denathio again. Where are you? Where do you want to be? Notice again the same five level score initial repeatable defined manage and optimized that you saw in this. This CMMI model has been universally adopted by many organizations and most importantly your non-IT managers are going to be familiar with it as well. Just put up here a very short list of things to avoid. Yes, buy-in but not commitment firing, trying to be overly complex, involving world hunger boiling the ocean. Goldilocks not too big, not too small, trying to get it just right. Overloading committees failing to implement not being able to deal with change management and they're assuming that technology alone is the answer and not building sustainable ongoing practices in order to do this. Again, all of these are going to be problematic if you're not able to do it. Finally, shadow systems are going to be problematic as well. Now, a couple of storytelling pieces here that are going to be useful. First one is a TED talk that we just love at Data Blueprint. Ted Sonak makes a great series of talks and one of them he talks about how great leaders inspire action and he notes that most people when they are very good at describing what they do. It makes perfectly good sense and it's a very reasonable thing that we should be good at it. However, we really do need to be informative describe more about how we do things rather than what we do. That's kind of the secret sauce but really the secret to his talk here is understanding that if you've got the motivational stuff down the why everything else follows in a very nice fashion. So really starting with the why first is a much better option than instead starting and describing what data governance is going to be. And if you have a burning bridge or an oopsie that the organization has gone through in recent years, then check it out and see what happens in that and where you should do better and how perhaps data governance might have avoided the particular oopsie. Give you a quick example on that. It falls into that category from a storytelling perspective. I was with a group of the defense department that was assigned to look at how Detroit made engines in excuse me not just engines but cars in the late 1980s and the idea was we were trying to learn from best practices of industry but one of the things we observed was that when engines were put on the assembly line and things were added to the engines success was defined as did it slow down the assembly line and that was really the wrong criteria because what that meant was you could use any bolt to attach anything to an engine which resulted then in different wrenches and different bolt inventories that had to be maintained as a result, maintenance of those engines was a much more complicated process. Toyota would stop at that level and then rethink and say not how many different bolts could be attached to it but how many bolts could use the same size bolt assembly and that gave them ideally one bolt inventory and one specific type of wrench which meant that their engines were easier to maintain and actually had better performance records than things that were coming out of Detroit during that decade. Again, a story that you can certainly analogize to many of the information and IT products that are developed by organizations today and I've seen a lot of organizations that could use improvement in that area. Another quick story, many of you are familiar with healthcare.gov of course the real problem there was that it had 55, not 55 contractors but 55 contracting organizations who were trying to work with it and anybody that knows anything about IT projects and realize that they were 6 weeks from launch and they still had not finalized certain requirements. This was absolutely a recipe for disaster. In fact, status group chairman Jim Johnson said anybody's written a line of code or built a system from the ground out can't be surprised or even mildly concerned that healthcare didn't work. It really would have been if it actually did work, the fact that it ended up working with success in and of itself. But interestingly here, another person who took a look at it, Marty Abbott who is one of the people who said help diagnose the problems with it he said it was pretty obvious from the first look that the system hadn't been designed to work correctly. Any single thing that slowed down would slow everything down. If you can imagine your motorized window in a car slowing your car down, that's the way the healthcare.gov was put in place. Another component to this too, it was a problem from a governance perspective. Many of the components were developed using traditional languages but some of them used big data technologies and when you tried to mix the two, what you ended up with was one part of the organization saying tell me what SQL to write and the other part of the organization saying no, no, no, we're not using SQL in here. SQL will not give you the results that you want to have. Now that may be a little too technical for some people but again governance would have prevented this from being again even an issue in there if they had had good governance around this. Another governance story that we did for helping the U.S. Army become better about this and the thing that was really wonderful here was that the Army basically governs everything and when we pointed out to them that the data was not part of their scope, they went oh my gosh there's something ungoverned, we got to fix that right away and that's exactly what they did. So we used their own culture to motivate them around that aspect of the process. Now the next thing that goes on here is that you see we did a project for the military where we were working with the suicide prevention project and this project obviously was one that was very near and dear to the hearts of people who were involved in it and we ended up having a governance issue where different people were doing different things on the project. Somebody would say I can use this data for this part over here and we had what we called a council of kernels that we put together on this and the council of kernels would sit down and try to figure out how they could make something work within certain constraints. We were very fortunate and had a senior Army official that we were able to bring into one of the meetings in there who brought a very heavy dose of management support and by the time in one of these meetings the third person had said you can use my data for this purpose. He stood up and put his attaché case on the table and made it absolutely something noise like this that got everybody's attention in the room. Remember all these people were kernels so they weren't able to argue with the individual in this case and said let's just get one thing straight here folks. We're here to help save the lives of the military and we're going to call all this data my data from this point onwards. Just that little change there absolutely empowered the team and changed from a can this be done to how are we going to accomplish this conversation. He also made it clear the mistakes would be tolerated along the way and we were able to build a workable prototype under these circumstances. Again the goal here was to save lives. People were thinking in one way but the application of governance in this case enabled people to really change. Now I've told this story dozens and dozens of times over the years. If I could get certain CEOs to adopt a similar position that the data belongs to the organization and not to the individual lines of business these organizations would have savings that could be measured in hundreds of millions of dollars able to do this. Anyway with our suicide story here what happened of course was that everybody pulled in place different patterns that they were able to put in place and we were able to look and save lives very very quickly. Here's another governance story just very quickly. One organization had trouble distinguishing between different types of tanks and so consequently we outlawed the use of the term tank in their organization. It was not precise enough how you're talking about tanks that fly, tanks that float that just simply didn't work for them and so we outlawed in business meetings the ability to simply say tank. We said you have to qualify it as a truck tank, a lawn mower filling tank, a boat tank, a tank on a plane in order to do this. And of course we also wanted to distinguish from other types of tanks in the process as well. Again another little bit on this but when certain parts of the military buy different types of vehicles they create literally millions and millions of data values. How many of those data values control the obsolescence of this? Well the answer was not very many and if you didn't know which ones you ended up with lots and lots of parts that were problematic and so we had one group that was able to identify $5 billion that didn't tie in between different types of that. Now a $5 billion discrepancy in inventory may not seem like a lot but that actually does come out to be quite a lot of things in here. Another little story and again these are just stories for you to use. Sorry I'm going to skip that but Dilbert here goes to do this. There we go. Spreadsheets of course are notorious for their problems in order to understand these things and when Mark Lees was getting raided by laymen brothers during the crash that they had there they actually had a spreadsheet reformatting error that caused them to buy all kinds of assets they hadn't planned on buying and the reason they did that was because a first year associate in the middle of the night reformatted the spreadsheet to make it look pretty and when they were still they bought 179 things that they hadn't intended to buy so you better believe at this point Mark Lees bank has very strong governance around their data I'm going to jump ahead I won't get to all of these stories now some of you can catch up with me some point in the future on them. Here's just one real quick one a really terrific one from British telecom here. They were trying to describe their way of going through. Now the story here was that British telecom had some challenges around their data and their governance organization needed a way of describing to everybody else how this new technique that they were going to apply in this case called master data management was going to in fact produce better results for them. That's a little really 45 second animation there cost them about 500 British telecoms to have it put together in flash but when it came from the lead official in the organization who was able then to articulate this very very well to the rest of the organization it became much much easier for everybody to understand what was going on there. Now if you want to copy that automation it's out there on our website you can take a look at it and download it they said they were happy to share that with everybody. To sort of come up with our takeaways here then data governance is a lot like Maslow's hierarchy of needs. While we like to do the fun things that are there those things really represent just the tip of the iceberg and that these foundational practices around governance are much more important starting off with a strategy and you can see governance is one of the key pillars that are related to that but that these foundational pieces are related by what we call a weak link in the chain method. That means that the entire foundation is only as strong as the weakest link. If data governance is a weak link in there you need to fix that in a way that's going to work before you're able to do this. Most organizations concentrate on technologies, data governance strategies focus on building organizational capabilities because absolutely the technologies will change over time and we're always asked the question you know can we do this without all that other stuff and the answer is yes of course you can do it without putting governance in place but it will take longer to deliver less and present greater risk to the organization that if instead you learn to crawl, walk and run your way to the very top. So from a takeaway data governance is increasing because of the volumes and because of the lack of practice improvement. If the relatively new discipline it must conform to existing constraints there is no one best way and you need to crawl your way forward in this. It's got to be driven by a strategy that complements the organizational strategy. Comparing frameworks can be very useful to help you get started on this. It directs then the efforts that are going on in data management and one thing I haven't stressed here but I should have stressed more in the presentation is that the language of data governance is in fact metadata. If you're not able to describe it in terms of metadata you will not be successful in terms of your program and that process improvements then can be used to improve these data governance practices around here. I've got two more slides on this. There is a Dilbert that goes into this. This is what you don't want your data governance to be understood as. The committee decided that the file naming convection will start with the date in the order of month, year and day and then a space, the temperature at the airport, the hat size is near a squirrel. It's supposed to be perfectly honest. It was a long meeting and we probably didn't do our best work towards the end of this. So be sure to not want your organizational data governance to fall into a trap here. You get the picture. I'm sure somebody put that together doing a very badly run data governance meeting and of course that's what we want you all to avoid doing all of that. Just to finish off the two presentation here I've got a lot of check lists and things at the very end that you will be getting from the downloads on this as well as a set of references on websites and books that are useful in there and now we get to the best part of the program which is where we get to answer your questions. So Megan I'll turn it back over to you. All right now it's time for Q&A. Time for you all to ask your questions so just click on the Q&A window feature at the top of your screen. You should be able to submit your questions through that Q&A window and we've had one coming already so let's go ahead and get started. The first question is what exactly well I've seen it everywhere from a director of data governance being reassigned to another part of the organization and the organization tearing up the old stuff and saying that didn't work we need to try it a second time to an existing organization that just says you know we tried to get that but we've really gotten some guidance from some people that could lead to some more immediate results so we've revised from a five year plan to a 90 day plan so really we've been at this long enough. I think I did my first rekindling data governance seminar in 2010 so now it's 2011 on that so we've had if you will recognize challenges in that area for at least five years with the reboot what you're really talking about here is that you know somebody puts in place this thing where you say we're going to do data governance what does that mean and somebody's got a very involved plan and the thing gets in the way of actually producing business results that's of course the last thing we want anybody to perceive data governance has is something that gets in the way of business results because executives aren't stupid they'll figure this stuff out pretty quickly and move into a posture that says we'll go back to the old way of doing it. Let me give you an example that Louis Broom who our CEO likes to tell apologies for Louis for stealing the story but he's not here today he worked with an organization at one point that at the very end of the process where they're making the bills that get to go out to all of their people who they're getting money from their customers they have a room full of 150 people and they sit down and every piece of data that is on the bill is examined and reexamined because large numbers of the pieces of data on there are wrong so they go through a process it takes them longer it costs a lot of money it's 150 people the bills don't go out their cash flow would be significantly improved sometimes the things are delayed by up to 30 days in order to do this and you're saying to yourself ah what a great case for data governance Peter right this is wonderful and Louis of course thought exactly the same thing and then they had figured out that this was the organization's most profitable quarter and most profitable year ever so management is saying to themselves okay if I'm making my most profitable quarter and the most money we've ever made by having a room full of 150 people maybe I should put 300 people in that room clearly here is an example of an organization that's getting the wrong message from its practices governance would be the process that would go in and say hey touching more than once for a bill is generally to be considered a bad thing and that if we can actually improve your cash flow and by the way the cash flow projections for this were astronomical I mean it was a very significant turnaround for this company by the way they didn't have a data governance organization at all in this organization so it was not a matter of rebooting in this case but we've got another organization that we work with where data governance has become a bad word it slows down development people asking questions and things like that and generally not being perceived as valuable so we don't want governance to be perceived as valuable and in those cases I do consider rebooting to be an appropriate option I hope that answers the question it's certainly a good one. Can you speak to any experiences where and how data governance reboot and organizational change management support and able each other? Well I think they are interdependent so I don't know for example where they haven't supported each other. One of the key functions that you'll realize if you're working in a governance situation is making sure the right data gets to the right people at the right time and there's no way of determining or managing that process unless you have good change management control practices. I don't think I could go quite as far as saying most governance efforts have failed because change management wasn't a part of it but I can tell you that every one that didn't have change management as a part of it failed. Again I hope that's real clear. So absolutely remember the language of data governance is metadata and metadata is about talking about what precisely we're putting in place. Again a horrible piece of metadata that many organizations unfortunately can say at this point in time is all of our data is of unknown quality. I would not want to be the person who had to deliver that message and if all of our data is of unknown quality then we have a collective problem and the only way we can start to put that in place to correct that is to in fact start through a good change management process. It's not just configuration management but it's also understanding some of the more again psychosomatic issues that go into it. How to get organizations to think differently about this process. Great. Our metadata start up. I've been using Oracle and the ability to create comments on tables, attributes, et cetera. I've been using code. I created a web version of a data dictionary. Is this an accessible start? Should there be more in the start up? I would say absolutely yes to give you the short answer to the question and unfortunately I bet about half the people on the call here do not understand what you just said. So make it read it again slowly. Just to meet around that because it's exactly what I'm talking about. So read the question and pause every second or so. It's a metadata start up. I've been using Oracle and the ability to create comments on tables, attributes, et cetera. Hold up right there. What they're doing is the callers questioning it and I think I know that this is too Peter. Good question. Thank you for it. What they're saying is they're going on in the specific storage locations with business terms to enhance them. They're making it more useful so it doesn't just say column one but in fact says person's first name used in this system. Next phrase, Megan. You're giving one second. Then using code. I created a web version of a data dictionary. So code is a freeware product for Oracle systems. It's a freeware piece of software that the individual has used to take and put in place some publicly available definitions so that if anybody wants to, they can look them up. So we've taken these definitions. We've enhanced the technical definitions with business terms. We've added some business terminology around it perhaps relating it to some of the business practices that occur and what they've done is created a metadata repository like functionality. All right. Go ahead. Is this an acceptable start? It's more than acceptable. It's a great way to start. The best thing that you can do next is to find a business sponsor who will use the information that you've produced in this sort of repository like functionality and say, okay, what if we extended the scope of this particular piece beyond the database that has been documented now to include data that travels between systems or across divisions. This executive would then be the person who would say, if I could achieve, and let's just pretend the number is a million dollars worth of savings. That's a large number for many organizations but it's not a large number in terms of being able to save 20 to 40 percent of your IT costs annually. Then you can go back out to the vendor and instead of saying, I'd like to buy a metadata repository with you about it, you can say, how does your repository improve on this business solution that I've already delivered? I can guarantee that salespeople aren't going to be able to answer that question. They're going to have to go to the next level of support, which is where you know you're having a really, really good engineer. We'll look up and see who submitted that question. That's an absolutely great question. I hope everybody understood that. That was kind of tough to do that visually thank you. The next question is what is the best intersect between BG and EIM or MDM programs? Do we need to establish BG before we start EIM master data management projects? Oh boy. A phenomenal question. If you guys will bear with me for a quick second, I'm going to pull up a chart here. This is not part of the presentation. We'll include it in the chart, however. The question that they're asking is, again, they're talking about interdependencies here. Here's the answer, which probably doesn't mean anything to anybody specifically right there. The question is what are the interdependencies between data governance and in this case they were asking about master data management. I illustrated it in this case with data quality as well. I was telling a different story using this chart, but if you look the way it works here, data governance makes the case and is responsible for data quality. Data quality is a necessary but insufficient prerequisite to the success of master data. The master data capabilities constrain the governance effectiveness in here. Here's one way of illustrating how all of these things are absolutely interrelated and dependent on each other, and your data governance organization instead of sitting through boring meetings making up the words to hotel California would actually be working on solving business problems by showing how if you implement one of these by itself it can help, but if you implement three of these in a nice integrated coordinated fashion, you can end up with much more than 3X results on this. An exponential increase in the effectiveness of this thing here. I didn't tell you guys I had planted that one, but I didn't, so I'm going to copy that out of that slide and paste it in here to the end of the presentation so you guys can have a copy of it. The next question is what government agencies are currently using or considering using the CMMI DMM model? Well, I can name a couple off the top of my head, but I don't know that it would be anywhere comprehensive. When you say government, do you mean state, local, or federal government? Let's see. I believe the interior department is looking at this. I believe Energy is looking at it. I believe that many parts of the Defense Department are looking at it, and I'm certain that is an absolutely incomplete list. Maybe Melanie if she, again, Melanie Mecca, if you Google her she will likely turn up on the internet pretty quickly for you. She could probably give you a more complete list. I think that, again, you will see virtually all of the government as well as industry going for this process improvement framework because it does represent a very simple and effective piece of intellectual thought that appeals to technical and non-technical people alike. Again, I hope that answers your question. If you run a comprehensive list again, please get in touch with me. I'll get you in touch with Melanie or you can try and get her direct. Megan. Yeah, definitely. Alright, the next question is, how did a data governance project lead to $5 billion in cost savings you showed in the presentation? Great question, and I did qualify that slightly. What we showed was a $5 million discrepancy, which is not exactly the same thing as cost savings, but it did add up to lots and lots of money. It was over $1 billion in savings. What happens is when you're managing large items and those large items have expiration dates on them, it costs a lot of money to store those large items and to maintain them and to take them out and test drive them and make sure they've got the right parts and equipment and type of oil and fuel and not to mention the weapons systems that are involved in that particular set of technologies that are there. And if they're obsolete, why would you spend any time doing them at all? You get the picture very quickly there. It's not just a matter of we're maintaining things that shouldn't be maintained. It'd be kind of like somebody saying, okay, we need to maintain all the biplanes because we might get attacked at some point. Not realizing that a biplane is not a technology that the Air Force uses. We've actually had some of the trucking companies that we've been working with occasionally will discover that parts of their fleet wouldn't be using them. And again, better data governance can lead to better management of those assets. And when those assets are better managed, we know to get rid of them, pass them on to somebody else, put them out in the desert, whatever it is you do with your obsolete equipment at the end of it. So thanks for the question in there. Again, it was not a $5 billion savings, but it was definitely a $5 billion mismatch between what they thought they had and what they actually had. What are the top points you would present to a state governor to put his support behind a data governance and data management initiative? I think that the most important thing to understand, and that's a really interesting question because we're doing some very interesting things here in the Commonwealth right at the moment. I've never been asked to do that before, but I would certainly say that the good, hard-working people in the various state agencies are doing the best they can given the circumstances, but at the most, without data governance, their efforts are coordinated at the workgroup level. And at the workgroup level, things can be effective, but if you take the things that are being done at the workgroup level and elevate them to the department or the agency level, they can be more effective and sometimes by several orders of magnitude in order to do that. I mentioned the book a little while ago. There actually are a couple of state government stories in the monetizing book. I'll put it back up here on the screen so you can see it. And they are actually real stories that talk about it. So the key points for the governor is that everybody manages the fiscal assets of the state in a responsible and an expectedly responsible fashion. People would not support, for example, the practice of anybody being able to spend money on anything. They would not be able to use any of their strengths by the legislature and by statute in order to do that. Data is also an asset. And in order to manage data as assets better, it needs to be governed. So again, we go back to our basic definition and say we're governing our data with some guidance. I would add one more piece on there, too, if I were presenting to somebody at a state executive level and the story, you know, may be relevant to something that had happened in the past or something that they were looking to happen to have happen and being able to do that. For example, if you're looking at a situation, I'll give you a very specific one that came from a good friend of mine who's probably not real happy that I tell this story but I think it's a wonderful story. She being a very data person got into an agency at one point and they handed her a data set and said, here, this data set sucks. And she being the kind of person to not like unquantifiable things said, I'm sorry, I don't know what sucks means. I assume it's generally not good but, you know, what do you mean by this data set sucks? And they said, well, just everybody's always talked about it and the quality is bad and since the quality is bad, the data set sucks. And she said, oh, okay. You don't know what sucks means. And being the person that she is, she took good use of her two master's degrees and went through and quantified the data and in fact found out that the data set that had previously been referred to as sucking was in fact 98% correct. Now, in her agency, she got immediate respect because literally this data set was known throughout the entire agency as the data set that sucked. And in just a couple of days' worth of good, hard work as a data scientist type of a process, she profiled the data and said, you know what, this data set is 98% correct. And everybody went, oh, what does that mean, it sucks or not? And she said, well, that's where it depends. You know, is 98% accurate enough? And again, this is not the example. But if you're transplanting organs, you can only get two of them wrong every 10. Excuse me, every 100. It may not be good, right? On the other hand, if you're talking about, you know, billing errors or rounding things and the previous error rate was 10%, 98%, actually sounds like a considerable improvement. So the real question that you're asking was, the governor's going to listen to my talk before and hear the bunch of theory. What the governor remembers will be the story about the data set being perceived as sucking versus now it's 98% accurate. And most governors like to quantify things. So I would always include a story of some sort that an individual like that could take away with them in order to really understand tangibly what does governance mean for their organization. Again, great question. Thank you. Ken, the next question is, what is meant by established data professional roles? It's in reference to a slide. It is. And we're just now starting to understand this. Again, another good question on this. Many organizations have now started to put in place professional categories for data managers. And some of them have the word stewardship in them, and most stewardship roles are associated with data governance. So if you look right now at the Bureau of Labor Statistics, they think that there are only two types of data professionals, DBAs and data administrators. Those titles were relevant in the 1980s. They're probably not relevant now. And so we do need to update the job categories to let people know that we have these new professional capabilities. This is what DAMA International has really focused on is to take these professional capabilities that organizations haven't known they needed and say, first of all, you need to make them into professional capabilities. And then what does it mean to be a data management professional in here? As you see, data governance is the center of the data management functions that are there. And if we aren't managing it professionally, then we are managing it by definition unprofessionally. And of course, we don't want to manage anything unprofessionally. And so each of the highways around here as well as the center now represent a new series of capabilities that we can now develop for professions. And most of these are not taught in college and universities. We're not here at Virginia Commonwealth University, but other than that and some others who teach in college universities, they're not part of any standard curriculum or even recognized at this point as being relevant subjects for teaching. Certainly not undergraduates and graduate students in here. So we are now starting to understand what these are and move them into a more professional discipline. And we are working with a group of labor statistics to enhance the breadth and depth of the job categories and how are you going to count them if they don't have a category for it. Again, gets back into, you know, can't manage it if you can't count it. That's a great question. Thank you for that. All right, it looks like that's all the questions we have for today. Thank you everyone for participating in today's event. We hope you've enjoyed it. Thanks again to Data Diversity and Shannon for hosting us. Once again, you will receive today's materials within the next two business days. We are looking forward to joining us for that as well. As always, feel free to contact us if you have any questions. Thanks everyone and have an awesome day. Shannon, did you internet stay up? Yes, it did. There we go. Thank you, Peter, for another fantastic presentation and thanks as always to our attendees who participate in everything that we do and ask such great questions. And Megan, thank you as well for facilitating. And as Megan said, I'll get the slides within the next two business days. There is also a request for a report that you mentioned, Peter. I'll get you that request, Megan, so that we can get that in the follow-up email as well. And I hope everyone has a great day. I'm sorry. Yeah, sure. All right, everyone, have a great day. Enjoy. May all your technology work. Yeah.