 Hello and welcome, my name is Shannon Kemp and I'm the Chief Digital Manager for Data Diversity. We'd like to thank you for joining today's Data Diversity Webinar, implementing the Data Maturity Model. It is the latest installment in the monthly series called Data Ed Online with Dr. Peter Akin, brought to you in partnership with Data Blueprint. Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. If you'd like to chat with us or with each other, we certainly encourage you to do so. Just click the chat icon in the upper right-hand corner for that feature. For questions, we'll be collecting them via the Q&A in the bottom right-hand corner of your screen. Or if you'd like to tweet, we encourage you to share highlights or questions via Twitter using hashtag Data Ed. To answer the most commonly asked questions, as always, we will send a follow-up email to all registrants within two business days containing links to the slides. And yes, we are recording and likewise, we'll send a link to the recording of this session, as well as any additional information requested throughout the webinar. Now let me introduce to you our speaker for today, Dr. Peter Akin. Peter is an internationally recognized data management thought leader. Many of you already know him or have seen him at conferences worldwide. In fact, he was just at Enterprise Data World. 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 11 books. The most recent is Your Data Strategy. Peter has experienced with more than 500 data management practices in 20 countries and consistently named as a top data management expert. Some of the most important and largest organizations in the world have sought out his and Data Blueprint's expertise. Peter has spent multi-year immersions with groups as diverse as the U.S. Department of Defense, Deutsche Bank, Nokia, Wells Fargo, the Commonwealth of Virginia, and Walmart. And with that, let me turn it over to Peter to get today's webinar started and to introduce today's guest speaker. Peter, hello and welcome. Thank you, Shannon. It's always a pleasure to be with you. And yes, we had a great conference. I think it was sold out, wasn't it? It was just the impact. I know that. We've had that, but now six different times with that one hotel. I can't ever remember that many people being there. It was just an enormous piece. Oh, Melanie, I keep passing past your slide. Anyway, it's my friend and colleague, Melanie Mecca, who's going to be joining me today. Melanie is the author of the material that you're going to be reading about today. But she and I go back a long ways in terms of our experience. She is the director of data management products and services at the CMMI Institute. If you don't know what the CMMI Institute is, it is the part of Carnegie Mellon that popularized the CMM framework on which this work is based. And she's got more than 30 years of implementing and developing strategies and data, just another wonderful person to work with in the data programs all the way around. But she is also the primary author of the DMM from day one. So what we're going to do today is run quickly through some of this. The original talk here was sort of introducing this, and it's been out for a while now. We're going to get a little further than that. So we're going to run through most of the first part of this in hopefully about 30 minutes. So we've spent the last 30 minutes talking about two specific use cases for this. The use case is a terrible word for this, but we're not going to mint that. You guys all know what we're talking about, how it is used in learning to get started on this. So let's just jump in and see where we go with respect to all of the, and I'll turn the first one over to you. Well, I'm very pleased to be here, and we're going to talk about how DMM is doing out there in the marketplace. And I would say it's making very good strides across a number of industries. And today we're going to present to you the fact that it's been adopted by the State of Arizona as a data management maturity framework, and we'll talk about how that's playing out. So the DMM is a measurement instrument. It's large and complex, and it has 25 process areas covering the fundamental data management practices that both the business and IT need to pay attention to to have an enterprise data management program, a very sound culture, and to have great data assets. It's architecture and technology, neutral industry independent. It's very quick. If you look at those little lights on the path there, you can easily find out exactly where you are on the data management capabilities map. So it can be used as a baseline for measuring your current state to immediately get the results to fix holes in the current program and come up with new initiatives for rapid progress. It's also excellent to engage stakeholders and have people enthusiastic about being data stewards. That's valuable, right? Okay. Next slide, please. Noyan, if I could just jump in there. You've been instrumental in helping rekindle a number of data governance programs that have kind of gotten solved because they weren't exactly sure where they were going. So the roadmap aspect of this is something that we've all found very, very valuable. Yes, because the governance motivation, first of all, we all work for Girl Scout and Boy Scout badges. We all like achievements. And when governance doesn't have a direction at the enterprise level, it often comes together and then the groups pass away with the end of a project. Whereas to have sustained governance and sustained interest on the part of the governance participants, you need to really know where the organization is headed. So the DMM can help you go from Samsara, the bottom where the elephant, angry elephant is chasing the frightened villager, all the way to the top of data management perfection where you're riding on the Buddha, you're now the Buddha, riding on the tame, the beautiful elephant in heaven. So we're all after the same thing, trusted data, lowering risk, cost reductions, additional efficiency, and additional standardization that makes the organization run better, and also regulatory compliance and competitive advantage. So the DMM is a very good foundation to help you do these things. So really key, we built on a fair amount of research that had been done, and this was the interesting part. I had a colleague named Bert Parker who worked for the MITRE Corporation that sort of came up with the initial piece. But people ask us all the time, we want to take our data management program to the next level. And of course, the first question is, where are you now? But if you can't measure it, how can you manage it effectively? If you can't manage what you can't measure, right? Very, very straight forward people. Absolutely. So if we were looking at this, we actually did this, I had a title, USDOD Reverse Engineering Program Manager back when I worked for the Department of Defense in the late 80s, early 90s. And we asked Carnegie Mellon to develop a framework for making things better, a process improvement framework. And the DOD funded the research in there. We went up to the SEI work with them. They responded with this integrated process and data improvement piece. And interestingly enough, the DOD required that the software engineering institute remove the data portion of the approach because their name was software, and they shouldn't be messing with data. Now, we of course don't agree with that in there. And this is what grew into the DMM. So there's a picture of Bert. He's passed on, unfortunately, but we did a great little piece. There is an article out there if you're really interested in reading the academic stuff. Or if you have trouble falling asleep, you can definitely get to that. The global figure copy of it at this point. Now, the SEI became a part of, sorry, was a part of Carnegie Mellon, but Carnegie Mellon actually spun you guys out. And so Mellon, that's the next step in the story, right? Now, let me... All right, I'll keep talking. Yes, we became the CMMI Institute. December 1st, 2012, and we brought a lot of the models from SEI with us, and among them the DMM, which was still under development. So we still have the CMMI Dev, which has just come out with version 2.0 being released around the world. We have a model for acquisition, for services delivery, people and talent, and of course, the DMM. And we provide dedicated training. We have partners and certification teams to support organizations and professionals who want to use our models and methods and grow in their knowledge. We are also now owned by ASACA. I was just going to tell everybody that I'm talking to you guys from Bogota, Columbia today, so I'm a little disobulated down here, where we started the data chapter. Melanie has just moved her household and her dogs, probably helping to participate in this as well. And they can have all joined in. What we are having is a slight technical issue with the refresh on my screen of the new slide. So if you notice any hesitations, audience, please be patient, because we're just dealing with slight issues on speed. So the CMMI Dev, and now we have 2.0 just released like a month ago, has been used by thousands of organizations, and partly because of its sensible approach, its practical intent, and its architecture, which is very easy to assimilate. And also appraisals giving overall levels, like level one, level two, level three, are done against the CMMI Dev. We did more than 2,000 of them in 2017, so it keeps growing. The real key for this though is that you'll see an awful lot of organizations that will want to start off their entire process with an assessment, and we think that's a very good idea. However, why would you use any assessment other than one that has this kind of background? This is a proven method. It has academic research that goes back over 20 years talking about how to improve it with really good feedback loops in there. So if somebody comes to you and says, I want to do an assessment or I'd like you to do an assessment, we really think that this is the way that you should go because you could be part of a larger process. The key to this, again, just go ahead. I was going to say, to get any statement, any requirement statement in the DMM, you had to go through the gauntlet of all objections, and we also had, in addition to our 50 contributing original authors, we also had 70 reviewers, Peter among them, Bill Inman, Peter Chan, inventor of entity relationship modeling. We had lots of appraisers with 30 years assessing organizations. So we put ourselves through the ringer to make it easier for you. They weren't afraid to get the hard feedback. So just in terms of results here, this is a bit that was done by the conference board in DC a couple years back, but nevertheless still showing that the CMMI approach outperforms ITEL, RUP, COVID, and even PMI in both on-budget and on-time delivery performances. And that is a really significant piece. Melanie's already mentioned here, again, these services offering the development, offering the acquire, acquisition, supply chain pieces, and the workforce development pieces on all of that. I guess we're also still working on the collaboration between DMM and the DIMBAC. That's a volunteer type of thing if anybody wants to get involved in that. We'd love to have some participation there. Let's just move on to the next one here. Melanie, back over to yours. All right. So I'm looking at the CMI model portfolio screen. Okay. So what we have tried to do with the Data Management Association, and I'm always on the road doing workshops and presentations for different data chapters, which I'm delighted to do to meet all these wonderful people, we're trying to eliminate any confusion between the two tools and highlight their complementarity. And they're highly complementary because the DMM is what you want to use if you want quick, precise answers of exactly how your organization is doing. And the DIMBAC is a fabulous reference resource for very useful technical implementation practical suggestions. So I think we can pass over the DMM slide because it's a very big thing. It has 414 practice statements and 596 example functional work products that are typically produced in doing those practices. I'll move it on up to the UR, which you do slide. Okay. I am still looking at the DMM. Sorry. Yeah, I'll just do this real quick. You can do a stop sharing, a quick sharing, and I'll catch you back up. Okay. So we emphasize behavior and its proactive behavioral changes on the part of both the business staff and IT that the DMM is intended to bring about. So you are what you do, essentially. And the image of the robo, which I cannot yet see on my screen, but I know is there, is essentially what it takes to have a great data management program. Yeah, Peter, we're not seeing... Go ahead. We're not seeing your slide. There we go. There it is. Okay. So it's important to be practical. It's important to get everybody rowing in the same direction. And the key to the measurement framework is that there were a series of levels created. By the way, when you're trying to tell this to your boss, if you just say it's the DMM for data, they'll get it. Because the process was really a very good process that was put together originally. Originally, the level one was that if you have a pulse, you're allowed to show up and participate in this. But if you get to level two, you've now got the ability to manage your processes, and then level three, define your processes. Level four, now you start to measure how your processes are doing, because only after you have them managed and then actually defined and measured, then can you get back to say, how can we do it better? So one concept has been the basis for TQM, ISO 9000, all sorts of other improvement frameworks that occur in here. And most everybody can get those live stages. Again, one point for a pulse, two points if you're managing that process, three points if that process is defined, four points if you are measuring things about that process, how are we doing? And then after we have all of that data together, then we can get together periodically and say, let's now try to make it even better. Yes, and of course that is the underlying principle of LEAN as well, Continuous Improvement, which is level five in the DMM. So the real key for all of this is that the DMM allows you to show the capabilities that you're doing and you can prove what you're doing in this. So again, you'll see the same five levels of performance here. I'm going to jump on to the DMM structure here so that everybody can take a look at how that looks. That should be 24 DMM structure. Okay, I'm looking at 23. Okay, so very quickly, in the DMM there's a great deal of material. So we have processes, we have introductions, we have questions. If you're an experienced data management professional and you really want to know how your organization is doing, just kind of do a finger in the wind by yourself, you can just go to the front of all of the chapters of the 25 process areas and pose those questions to yourself. And it'll basically tell you, the questions are aimed at level three, green, good, go, enterprise-wide. So if you're answering yes to all those questions, then your organization is quite mature. When we score against the DMM, we are only scoring two things. One, the functional practices in the process area, and if you achieve either a 2.0, where you've satisfied all those processes, or a 3.0, then we also apply the maturity practices, known as generic practices in CMMI, and we call them infrastructure support practices, meaning do you have a policy? Do you have resources? Are you tracking, monitoring, and planning the process, et cetera? Sensible housekeeping to make sure that the processes that you put in place are resilient and can withstand any kind of stress to the organization. The other part of it is it's actually pretty easy to learn, too. That's important because people will then look at a diagram such as you see on the screen here now. It looks a little intimidating at first. This represents the upward theoretical complexity around the entire process. What we're really talking about is data management strategy, the ability to manage the data coherently, the ability to manage data assets as a professional, a class of asset management professionals. Make sure that the data is fit for purpose and sufficient quality that we have to make sure that you're doing it with the right process and the right technologies as well. And of course you need some supporting processes in order to keep all that rolling. What I usually like to say is that data is forever when your organization opened its doors for the first time. It started collecting data, and until the day it no longer ceases to exist, it will be collecting, storing, and using and producing data. So data is everything. It is the ground on which you walk, and you need to do a good job with it for your own good, for your mission and business processes. You will no longer need a data program and you no longer need an HR program. That's right, or finance. So data management... These are the five areas, right? Yep, that's right. So I'm just going to say, you know, just kind of summarize them in a sentence or so. Data management strategy is the reason why enterprise data programs usually either do not exist or are kind of half-baked, because they're not starting with a top-down view of what you want to achieve in the strategy. So we have the strategy itself, so we give helpful practices and kind of a template outline for how you can do a strategy. We talk about the constant importance of communications across the entire life cycle of the data management program, because there's a lot going on. You need to communicate from governance and so on back and forth to the data management organization about the program. We talk about the importance of the data management function, whether it's a centralized organization. Then we deal with governance. Governance as such, I just want to say here, that governance is everywhere throughout this model. About 105 of the 414 statements are either specifically using the word governance or talking about agreements and approvals. So governance is built in. We're dealing with the governance activities within the 25 process area that we cover. So we don't show governance as central as the DIMBOK-1 does, but governance diffuses every single thing that you do in data management. We put a lot of emphasis on business glossary as a part of metadata management because the business staff are critical to get the business terminology right. That's why we separate it from the rest of metadata management. Data quality is a 360-degree program starting from a top-down strategy. If quality is so important to you, why haven't you done a plan? That's basically a way you could look at it. The big three are governance, architecture, and quality. Those are the three big pillars, broad pillars of data management. So quality is where the user gets the benefit from the good data. We, of course, talk about profiling, business-driven assessment of quality, and good practices for cleansing data and lowering costs for data cleansing. And then there's data operations, which is essentially doing a great job on data requirements. You know that most organizations do a wonderful job with functional requirements, and yet they neglect the data requirements. So many organizations supplement their documents with templates and other things that will help them get the data straight. From the beginning of the project, please. Not in the design phase in a hurry with a couple of developers and a quick typist. That's not the way to do it. Data life cycle management, of course, is tracing data end-to-end from the source through the landing places through the end destinations. You don't need to do that for all the data, but for the critical data, you should get that done. And data provider management. I just wrote an article about this for TDAN. Data provider management is monitoring and putting quality requirements on the data that you buy or acquire externally or from inside the organization. And if you, as a data management professional, want to make a big splash in your organization and save a lot of money, suggest that you analyze all the external data feeds. One organization did this recently and saved 2.3 million a year through discontinuing old feeds and changing licenses and so on, and got better quality from the vendors. Platform, what do you build? You should involve the business. You need standards. You need to make sure that you pick a data management platform, which is one or more big systems that are going to manage the enterprise data. You want to make sure that that meets business needs. Peter, I know you can tell them about many, many, many builder-buy mistakes that organizations have made. I know of several that are multimillion-dollar errors, so please involve the relevant stakeholders. Best practices for data integration, and then we talk about historical data, data archiving, and retention. And I will tell you that most organizations do not have business rules for records retention, unless they have a big central program that they have to do it. But it is very useful because, as you know, even in today's world with very fast processing, if you clutter your production system with inactive, aged, or bad data, your performance is going to suffer. And then we have good housekeeping practices. How do you measure? How do you do good metrics? And how do you apply them to data management projects and programs? How do you manage the process assets? Do I know where to find the data profiling template? Do I know how my business term process is going? Has everyone followed the steps? Can I measure risk in the data projects and programs? And do I know which version of the enterprise data model is current and should be used? So sensible processes that help you do a great job. These are very common in the IT development environment and very uncommon in data management. So we just want organizations to think about applying them there. What you see is a very nice framework that has been developed that describes five data management practice areas plus the supporting processes as Melanie said. And then we can rate them each on a one to five scale. When you put them together, you can end up with results such as this. This is what I did for the insurance industry a couple of years back. Notice we're not naming any specific insurance companies, but we are able to say that the general level of their performance in their areas is uniformly low, and they could all benefit from doing better at data management in the insurance industry here. Another way to look at it is for an individual company. This was an airline, actually, that we did. And the interesting part about the airline was that the executives were kind of staring at me as I was presenting this thing. So it was ones and twos and what does all this mean? And I said, well, here's where it gets really interesting. Right now what you're seeing is your performance. And then when I put these red lines up on the screen, now you're looking at your competition. And that got the executives pretty quickly. They were attentive. We could also compare them to all respondents. So once again, this unified process, the unifying framework, the fact that she did 1,900 assessments last year alone tells us that we can now start to incorporate this at a much higher level of organizational readiness. And of course, the key for all of this is that we have to make the areas that are not performing well come up to the right levels. So make that a little bit better. The foundation for all data practices, anything that you're doing out there, is only as strong as the weakest link. In this instance here for this airline, we are seeing three weak links, the three ones that are in the program. So they could pour literally a billion dollars into data quality, and they would not improve their overall score because they need to have the rest of this come back out and take a look at this. And as we didn't tell you which airline or which insurance companies, but the World Bank told us we could use their marks on this. And the interesting part about this was that we did it for three different groups. In the purple up at the top there, that was their treasury function. In the sort of teal color, that's their information systems group. And IFC is their business function. So the business function was performing at level fours. These were the first level fours that we had ever received when we were doing this. And what it told us was that these folks did not need to go outside and hire external consultants. They needed to walk down the hall and find out what the rest of the company was doing. Again, we can put on here some other managers just looking at industry benchmarks and overall benchmarks. Something like this is invaluable to the organization because it says you're doing things good internally. Let's celebrate the internal successes and make sure that we can push them further and faster out all the way around. I haven't done one of these obviously since 2012, but this is what we call the era of big data. And the overall stats for the entire time are just completely unchanged. They are statistically not different. Uniformly, we are not improving. I think we know that the amount of data is considerably improving. We've used this in the state of Virginia in particular to help the various state agencies start to get better with their own internal practices. And we're finishing up a study that we've done at VCU recently with my colleague, Joseph Hola, who has been doing the state agencies for the past three or four years and will be able to give you a comprehensive look at a state-level piece of that as well. But what we're going to do now is turn it over to a couple of specific use cases that Melanie has been working on and that we've been involved with as well around the state of Arizona in the first one. We'll see if we have time to get to the other one here. So I'll just sort of stay tuned on that one. Melanie, go ahead and tell us how we've been going into Arizona. I think it was Jeff had set up, Jeff Lockloff, who had been sort of an interested party in data and he had gotten some money set aside for him. He really didn't have a position yet, but he called everybody together for a big conference and he developed a plan and what he's doing now is actually executing the plan in a very, very systematic fashion. Yes, Jeff has had a very... He's the state data architect and he's had a very embracing vision of raising the state of data management across all of the agencies in the entire state of Arizona. He's with the Department of Administration, the Strategic Technology Group and he is concentrating primarily on policies and he also is the point person for a five-year indefinite quantity and definite delivery contract, of which we have three bidders and three accepted bidders, CMMI Institute, Data Blueprint, my fellow speaker, and CenturyLink, who does business intelligence implementation. So we're all working to help Jeff get this vision realized. And on the right, you can see a sample from his data governance operations policy. Notice there if you... I'm sorry about that. Yep. You're good. That's all right. They did see it. They're still being... So DMM is providing metrics for them? Oh, we went back. Okay. For those of you who joined late, we are having a little issue with the speed of screen refresh. So if you hear us hesitate, it's because we are waiting on our next slide that hasn't quite given its vision. But Arizona is a wonderful laboratory that can be applied to large organizations. We've had some federal organizations interested in what's going on in the state of Arizona, and also any large organization or globally distributed organization can benefit from what Arizona is doing. So we, with Data Blueprint, helped Jeff put on his second annual data management conference in the spring. And then we also did a course, our building enterprise data management capabilities course, where we had people from many different organizations. So Arizona, at the governor's office level, has a massive lean transformation effort called the Arizona management system. And that is essentially what they are doing to improve performance and customer service for all of their clients, citizens, residents, receivers of services across 100 agencies. So we had 11 agencies in our class, and as a result of that, we are doing a number of assessments. Against the DMM, we have done four of them now. So the students were interested in more training, and they also, Jeff had used these groups of willing participants to help him write the governance policy, and now they are helping him write other policies like they are working to finalize an interoperability policy, which as you on the line know, that's huge because you are standardizing data from agencies that came to be at different times and that are quite different from one another. So it's a big, big task in any state or any diverse organization. And the DMM is helping. Go ahead. The flat team piece is a particularly innovative approach, which is the idea that, you know, it would be nice if banks could work that way but they can't because that's called collusion. But in the state government, it is wonderful to have people working back and forth across agencies and helping each other out, because I know it's going to make all of their collective jobs easier. And so the flat team concept is something that I think everybody is very excited about. Yes, and he's very fortunate to have approximately 20 stakeholders representing these various agencies working with him on these policies, which is wonderful. We would certainly recommend that approach to any state because everyone here who's speaking today is a big collaborator. That's how things get done. So I think the Department of Corrections, I was going to talk about, okay. So when do we use the DMM? When is it actually useful? And of course, it won't be surprising to you to hear me say that it's useful in almost every circumstance, but not just for enterprise-wide programs, but if you are about to do an architecture redesign, you definitely want to see where your data management practices are. For example, Fannie Mae's Enterprise Data Architecture Effort, which has been going on for seven to 10 years, a great part of that architecture effort was not just the design of all the new data stores to replace the Enterprise Data Warehouse Layer, but a determination of what they needed to serve it in terms of data governance, in terms of data quality, in terms of metadata, in terms of requirements. And I will say on their behalf that Fannie Mae is probably the strongest organization overall in data management that I have yet encountered in the last 20 years. So they've spent a lot of time, money, and expertise in their great shape. Go ahead. I was going to say it's kind of interesting because both of us do a lot of work with the federal government. And one of the things we keep hearing is that the federal government needs to go learn from industry. There are certainly some things that they can, but it turns out from a gross management effort, the federal government is actually slightly ahead of the private sector. And it's probably because it's some of the things that Melanie's talking about here where you can do this behind-the-scenes coordination. Everybody can get involved and work in these areas to share their expertise whereas you just can't do that in the private sector. So it's a really, really fun thing. And we do like to celebrate when the federal and the state governments actually do outperform the private sector here. Yep. I'm going to give you two other examples of that. When I was working with FINRA in the 90s to help them come up with logical and physical data modeling standards, you know where I got my base material? The Environmental Protection Agency that already had really good standards. And we got permission to adapt that. So that just shows you about the government and also many agencies, including the Pension Benefits Guarantee Corporation, where I worked in enterprise architecture, they have a working third normal form enterprise data model for all of their fundamental data, and they require new projects and redesigns to start with the enterprise data model and add to it. So advanced practices do happen in the government and have been happening for a long time. So anyway, there's a lot more that you can do with the DMM. I mean, if you want to know what your data quality program should be, let's find out if it's supported by your governance and by your metadata and everything else. Let's go to the next slide. One thing this tells you, too, is that very often organizations simply aren't ready for the technology that they're accepting. My analogy is you probably aren't going to expect good results if you hand the keys to a branded Tesla to a 16-year-old that has never had any driving experience. Just call me crazy, but I just don't see that to be a good outcome there. And I think in many ways the way technology is sold is that they forget to tell you that there's a learning curve involved and that there are ways of achieving that and that there are ways of telling when your organization is ready to jump into some of these advanced technologies that we have here. Yes, and as Peter has mentioned about big data, I've been on a quest for many years to see if these fundamental practices that happen to be in the DMM, is there anything about the end platform of big data that would cause you to add or change any of these practices? And so far, after talking to many, many big data experts and vendors, the answer is no. The data still has to be in good shape and it still needs to be managed. The challenges of big data are how do you structure, how do you index, and how do you maximize performance. But everything, if you have no metadata, then your big data platform becomes just an even bigger data swamp and that's why organizations are unhappy. Because as you know, vendor will never tell you that the capabilities will not solve the challenges that are not technical challenges. That's why Peter and I, among many others, are advocates for do the sensible thing first, plan, and know what you're getting and know what you're implementing. Don't get us wrong. We love draining the data swamps that are out there, but it's really not a very productive effort for your organization, so absolutely. Yes. So we teach and certify against a specific DMM assessment method that was actually developed along with the development of the model, because the model is big and dense and you don't want to spend six months doing an assessment. There's no money for that kind of thing anymore in today's world, whether it's in a government agency or in the private sector. So what we do is we work essentially with a broad range of stakeholders from all of the affected business groups, as well as key players in the data management organization, if there is one, and key program knowledge people from IT who understand their systems and data stores. So we put them all in one room, or some of them, of course, are on videoconference, depending on the organization. We go through the entire model with them. We give examples, we facilitate, we get the consensus on this, and then we look at work products, and then we do supplementary interviews with key players or executives who have specific needs and want to get out of this. And that all takes four intense days. And then essentially the report from the DMM is fairly easy to write bottom up, because you have notes on the scoring mechanism, you provide the scores, what your conclusions are about those scores, specific improvements that they can make to raise their scores and on certain practices. We also find out their strengths as an organization, how their organization is put together. We are able to give them organizational themes that we've observed that impact on data management. And finally, we end up with a slate of initiatives that are like many business cases, typically 8 to 12 projects designed to rapidly accelerate their capability and maturity according to the DMM, but also to meet their business needs. And these business needs are different for every single organization. Sometimes organizations are interested in doing this and they ask us or our certified partners, can you give us a sample? And we have to say no, because although the DMM is the same for everybody, everyone's results are completely different plus we sign NDAs. You don't always want your competitors to know that you didn't score very high against a benchmark of any kind, right? So we don't do that. A summary slide though, this is a very high level overview of what you're looking at with this. But what it does tell you is specifically observable specific behaviors that your organization is doing. There's no interpretation on this, it's fact-based. And that's one of the most wonderful aspects of the entire process. Yes, I really like spider charts because it says everything in one visual. So this organization obviously spent a lot of time in metrics and for whatever reason they've done a great job there. And if you look down at historical data retention and archiving, they probably have a good record retention policy and archiving rules. And if you look at data quality strategy, it looks like they've planned it quite well. But then if you look at the rest of data quality, it hasn't executed yet, right? And so, and provider management, that could mean that they have done a super job on controlling their data sources. They have interface control documents. They have data quality requirements for Bloomberg, et cetera. So this is not that, I mean, it's a little strange. This is not a real organization. But nonetheless, it does illustrate that people in organizations pay attention to things based on time and resources and priority. So big projects will move the needle in one area. So let's say I'm redoing my data warehouse. I'm going to need to update my metadata for all the data that's in that data warehouse. So then you would see a lot of practice improvement in metadata management. But they might not have paid any attention to the target architecture other than the warehouse. So that might be a low score in data standards, et cetera. So everyone grows according to time and attention. And one of the overall uses of the DMM is to get people to realize in organizations that, as Peter pointed out, the weakest link will come up and bite you over time. So if you have your end money or things like that, it's very clear to go in and hit some of those pieces. And again, Melanie was talking specifically about the data quality area, which is the sort of light blue between the purple and the red and the bottom right-hand quadrant of the diagram. Got that great strategy, but they need to get work in this case of doing some profile assessing the quality and actually cleansing the data. Because the plan is good, but you need more than just the plan in order to get forward. So this is what the assessment gives you at a very high level. It also, of course, gives you lots and lots of detail below it. Most management doesn't want to sit around. They just want to look at this and go, okay, great, how much can you fix by Friday, right, Melanie? That's right. And sometimes the circumstances and the aspirations and challenges of a particular organization allow us to go the extra mile. For one organization, we basically said that they were planning on doing a client MDM project. And then they said, well, then we want to use Salesforce for the rest of it. And we realized that there was a missing piece in the architecture, and that was essentially customer relationship management. They didn't have a reliable, well-architected contact system. So they weren't able to provide the proper customer service, and they would never, never, never, without that they would never get the 360 degree view of the customer that they eventually wanted. So whatever we see based on the DMM, we will tell them. And typically these projects that we recommend typically get prioritized, funded, and they start them. And if the organization is not very capable, like if they don't have any data management organization, then we basically say, this is the data management organization you should start with, and without a data management organization, you really will never get beyond level two. And we give them a structure and everything. How to get going. This benchmark slide essentially is showing over about 20 organizations that we and our partners have worked with what the average scores look like along the range of highest score to lowest scores represented by the blue bars. So if you just look at the, if you look at the diamonds that are mid-range, you know, in all those score ranges, you can see that data management, you know, and we're doing this cumulatively with every organization we or our partners work with, we're updating this benchmark. So it's all over the map, right? Certain things they don't do a good job at. Very few organizations do a good job at, like data life cycle management in the middle, sources to target tracing. It's a lot of work. You should have done it over the last 20 years, but you haven't. So now you have rework to do and it's tiresome and you only do it if you have a regulatory or really keen competitive purpose. And then in other areas, governance organizations are usually doing pretty well. See, they're above three. Now, our average on the high end, I will say is skewed by several strong organizations, but primarily two. One is the Federal Reserve Statistics function across all 12 banks. So they are on track with their long-term data strategy on budget and everything is going extremely well for them. And of course, their job is data and has been for 40-some years. So, you know, they deserve the credit that they got on their assessment. And I already mentioned Fannie Mae, who has spent loving care, time attention, and then lots of money on their architecture and data management program is a star as well. And there are others who are quite strong in certain areas. So the really high scores are basically those two organizations. And while they may have skewed this a little bit, this is still, the point of doing it this way is that you can compare your results against national averages. And as we get more data into here, we can actually compare them against industry sectors. So if you're in the insurance industry, we can compare your company against other companies in the insurance industry. Again, we keep everything completely confidential. We don't want, you know, to be ratting people out on this, but it is very interesting to see how these things are evolving. And again, if you have the alternative to go and just do a random assessment, you know, it doesn't tell you anything. It's somebody's idea of what might be working as opposed to this 20-year history of academically funded research and very, very robust results in this case. So now we're onto the training and the certification pieces on that. Oh yes, very quickly. Yes, very quickly. Okay, so essentially we have a three-day instructor-led class where we come to your site and it is a very highly interactive class focused on implementation. So even though we don't get into the details of the statements in the model, we do give you a beautiful printed copy of the model and we refer to the model, but we're talking about why organizations should strive to be better in these process areas and what are the challenges to implementation and what are the success factors that help you implement quicker and better. So that's what that's about. We also have, without the person instructors and the exercises, we have a web-based version of that. So eight to 10 hours, it's inexpensive and it's essentially a way to fix the perspective of enterprise data management in your mind forever. Inoculating you from any, you know, falling out of perspective. You'll never be limited to just one business area, just your IT point of view again if you take that course. That's my sales pitch. We also have two more intense courses. One of them is primarily focused on using the model for consulting and the other is focused on doing assessments for any organization with our method which is very tightly wrapped and very structured to help you do a fabulous job right out of the gate. And we have another assessment coming out that is the data management associate assessment and that's going to be a 60 question exam very similar to a CDMP module. Well, while you're saying commercial, you know, I want to make sure that we just do take that Isoca is not an industry support group. It's not like you guys are out there like the other vendors trying to compete in this particular thing. This is done for the good of the organizations, for the good of our entire industry in the back. That's exactly right. And we've been working with some products with Isoca including a mapping to the COVID methodology and we have other products in line to work with them. So back to Arizona very quickly. I already talked about the Arizona management system which is a massive effort and we were so extremely impressed with the agencies who were working with AMS. I mean, they really got their act together on the business process side and they came up with lots of improvements and they saved money and it's getting really very, it's a very good thing to do, the lean re-envisioning of state government. So we're doing a course now but I have a slide on that later, but I'll just briefly say it's a two and a half hour course about being a data steward and we're working with KIK Consulting on that and we have completed the course and it will be put up by the computer-based training department of the Administration Division of Arizona and it's going to be available to 20,000 people. So let's kind of go down to that slide, Peter. Okay, multi-year EDM program or different? This is basically what they are doing. Well, all right, I'm sorry. I will interrupt my narrative to just mention that we talked about this already. So we're working to train. Yeah, we actually did that one too. So this was back to Jeff's piece. There we go. All right, there we go. Okay. No, we did all that. Okay, we're good. All right, so I just want to talk a little bit about, so our course is going to be offered to 20,000 staff members and we worked as hard as we possibly could on this course to make it compressed because he said no more than two hours and we're like, oh my God, you know, our course is just five days long. You can't possibly get anything meaningful in two hours. But we have managed to distill a great number of activities and also my favorite part, the analytical thinking skills that a data steward needs to learn to love in order to get reward and joy out of the job. So we talk about data quality dimensions. We talk about how to define business terms. We talk about working in data working groups. We get into some detail about how to help your agency make decisions about acquiring data, how to do data requirements, and the course is supported by 10 detailed templates for how to do things like a data profiling template. If you're going to do a data profiling effort and you as the data steward are working with IT, how are you going to plan that? How are you going to scope the data set? We give you a bunch of information and questions and material that you can quickly work together a plan and a results report. So that's the kind of thing. Very, very practical. So I'm on your next step slide. Is that the right one? Yes. So we will be doing additional training in Arizona and a lot of the people in our first training are getting their EDM associate certification now. We have done assessments for the Department of Corrections and that was just thrilling because we got to apply the DMM to a prison system. That was definitely a different industry and they were extremely impressive as an agency and they were really wanting to become even more data-driven than they already are. We also worked with the Department of Water Resources and Lisa Williams, the Enterprise Data Management Manager there is doing a presentation with us at DGIQ on how she set up her entire data management strategy. So that's going to be a wonderful one. And we worked with the Medicaid agency, the healthcare cost containment system, and that was really fascinating as well. And when we were walking out of the executive briefing with access, their director and the deputy directors were saying, okay, which project should we do first? They were already getting ready to kick these off and that made us very happy. And we've just finished an assessment, I mean like Sunday, finished an assessment report for the Department of Economic Security that handles things like food stamps, child support, services to the elderly. It has many missions. So with each of these, more and more people are educated about data management and more and more agencies have, like the DES that I just mentioned, are going to put all of these recommendations into their data strategy and roadmap for the next three to five years. And they've set a target. They've set a target for level three, as well as many other agencies. So it's excellent to use a benchmark. Even if we did write it ourselves, we have to say that using a benchmark is a strategic approach for bringing up the quality of data management and sharing in a state. And to our knowledge, this is the first state that has actually set specific targets for agencies to achieve certain levels within certain years. And that's just a huge, huge advantage. We've got about seven minutes left, Mellie. Let me jump into the CP. Please. The other part of this. This is really exciting too, and I don't want to miss this opportunity to get this out here. So tell us about how Health and Human Services has helped participate in this. Okay. And I'm going to go really fast, and hopefully my screen refresh will be as fast as the leaders change slide button. So mismatch patient data, duplicates and overlays is the third leading cause of preventable deaths in the US. And 86% of people know of someone, 86% of healthcare providers know of a medical problem that was caused by a duplicate patient record. So the Health and Human Services Office of the National Coordinator for Health IT was looking for a data management framework to start building their own profile for patient demographic data quality. They chose the DMM. And from this slide that you can see that there are a lot of problems and errors that can be caused by duplicates. So if you want, yes. So basically you wanna decrease the operational risk and the risk to the patient. Patient safety is paramount. And in an ancillary but also important way, claims and billing are protected by lack of duplicates. And of course, interoperability is helped by that. Interoperability in the healthcare space is a tremendous challenge. And organizations and organizational industry associations are dealing with this all the time. Different vendors, different health information exchanges, the electronic health record. So they selected the DMM because it was behavioral, behavioral, fact-based and practical. So what we did was work with the ONC and work with some major publications that they had put out where they had interviewed and surveyed really tens of organizational elements providing healthcare, big healthcare systems, small ambulatory practices. So they published a huge research study. We use that as our primary source. And what we did was take the DMM and slim it down greatly just to cover as a nice tight-fitting glove the scope of patient demographic data. So essentially, we made a profile for master data. In this case, patient master data. Patient identification and matching final report was a primary source. So instead of 25 process areas in the DMM, we had 19. And instead of 414 statements, we have 76 questions. We turned them into questions because we wanted more discussion. One of the things that's very important is that across the patient care life cycle, starting with registration and ending in billing and maybe even collections, all of the patient care areas are very mission-focused on what they provide, whether it's direct primary care, laboratory pharmacy, et cetera. But they do not talk to each other very often and patient demographic data can be modified or changed at almost any point in patient life cycle. So we wanted them all to get together in the practices. We changed the levels to tiers because we want even the smallest practices to get started and be able to get a good score once they're organization-wide. And we slimmed down the text, the explanatory text and examples significantly. So the entire thing from front to back is 96 pages and most of that is contextual examples and suggestions for implementation. So data standards at the bottom in the box there, data standards is what's going to fix this eventually. So the national patient ID has been around forever, but it has maybe a better chance of passing since there was a Kentucky senator who withdrew his opposition to it. So that would help a great deal. But in the meantime, within an organization and within a health information exchange, you can improve the quality of the data significantly through using the PDDQ. And I think what this really shows is that the work that was done on the original DMM is in fact at the right level of abstraction such that if somebody wanted to come to us and say, let's do this for another industry or another healthcare group or another entirely different place, it seems to work well. So we think that they're going to be more of this type of customization and specialization, particularly as opposed to less of it. Right now, we're approaching the ability to do a DMM-based profile mapping to the general data protection regulation. And this will also give us an opportunity to add new process areas for the future DMM 2.0, and that is identity management, PII, entitlement and permissioning, data classification, et cetera. So that's going to be quite crucial. Go ahead. Yep. Oh, I was just going to say, the other thing that the HHS wanted was they wanted one responsible person. So the data quality coordinator, whether it's a person leading a large group in a big healthcare system like ETNA or whether it is a couple hours a week from one part-time registrar in a very small practice, they want a designated individual to be responsible for administering the PDDQ and monitoring progress and improving quality. And this was piloted by Kaiser Permanente with Ocean Healthcare System of Community Clinics in Oregon, and they used a couple of process areas and they found that they were essentially improving by 25 to 30%. That is lowering the number of duplicates. This is just a sample of some questions in one process area data quality planning. So tier one is, have you involved the people in the different areas of the organization? Do you have a plan to improve data quality? That is, but notice we say through quality rules, which takes a little more work. We have suggested work products for them to work on. At tier two, we need to see this spread further across the organization. We're looking for, even if it's one page each, a policy, some processes and some guidelines to improve data quality. And then we're looking for, are you now monitoring your plan to see that you're actually doing it? And finally, it would be all buttoned up in tier three where they would have defined roles and responsibilities and accountability for data quality. So this doesn't have any of the explanatory text. There's one example of explanatory text. So this is part of the, for 2.1 in data profiling, we're basically saying, here's what a profiling method is, here's how you do it. And here's our favorites chart. Again, this is a highly performing organization. We're looking for everybody to be level three. That's HHS's target for healthcare organization. So we've taken you on a sort of a rough line, and I apologize for the technology. It's not reflecting quite as fast as you said, but there is a way, when someone comes to you and says we want to improve our data management capabilities, there is now what we consider to be an authoritative standard. And I'm not saying that because we use it. I'm saying it because it has a 20-year research record. Put any other assessment up against that. And I think they all fall short. The effort that's gone into this has been community-wide. We've done a tremendous job. Melanie has been fantastic about involving all kinds of expertise in the community to make sure this comes out the way we'd like it to. And that we are seeing significant traction in the marketplace here. So hopefully this is giving you guys an idea of what sort of things we have. Melanie, I'll leave a few minutes for you to conclude here, and then we'll move it over back to Shannon for the Q&A portion. So I think where to next, as I said, we're turning our attention to data privacy and security. And we are still wanting to, we probably will be doing another course for Arizona. Jeff Walkover, a client, is interested in a course for data managers, that is, people who are setting up and monitoring data governance. And he's also interested in a course for data owners. So if you are the business sponsor of an information system or the IT program manager for an information system, what are the skills and knowledge and techniques that you need to know to be great at your job? So we're really hoping to build a whole suite of training for that state. And we also are speaking to the Shanghai Data Exchange in China. They're very interested in the DMM in it, and hopefully we'll be doing a couple of classes there and translating the DMM into Chinese and possibly working with a major university in China. So that would be great. And we also are very interested in translating the DMM into other languages. Spanish is probably on the horizon very quickly, since we're getting more traction in South America and Latin America. What are the reasons I'm here in Bogota? So let's turn it back over to Shannon and see what sorts of questions Melanie's got to learn more. Pages will include that, of course, with the documentation. But what else can we tell you all about this exciting development in the world of data management? I love it. Thank you, Peter. And Melanie, thank you for joining us. Always a pleasure to have you join us. And just to answer the most commonly asked questions, I will be sending out a follow-up email by end of Thursday for this presentation with links to the slides and links to the recording of this session. So just diving right in here to the questions, where can I find a standard data profiling template? Well, I mentioned that and that we are providing that to the state of Arizona. And if someone wants to email me and ask me very nicely, I might just give it to you. It's something I'm working on from a number of profiling efforts that I planned and managed in the past. I love it. I'll be sure to send that out. And so what is the difference between DMM version 1 and 2? OK, so DMM only has version 1 at this point, so there is no version 2 yet. We're planning to start that next year. So that's the answer to that one. That's an easy answer. Well, which factor from the organization do you think is key for DMM implementation to succeed? Executive commitments at the highest level. Because if you are a business data expert or you are a business manager in a business unit, you are focused on your doorstep and sweeping it clean and improving it and painting it. I'm putting plants on it, right? You're not focused on the entire block. So it's very, very important that the executives be committed to managing data as an enterprise asset. And once they are, then they can help give juice to this effort to spend some time and money on it. And I'll add on to that too, Melanie, that it's actually quite important too that you involve some aspect of finance in here as well. It's great to have executive management. In fact, it's insensible with what you're attempting to do. But if you can tie this to some very specific actionable outcomes, it makes it even more important. So one of the things we did here in the state of Virginia was to look specifically at how agencies were doing their various internal practices. And we found that with better managed data in one of the agencies, they were able to take over $1 million very easily from what had been in the previous incarnation of the process and intake process. So they were asking lots of questions when they were running to a person who needed goods or services. And the questions turned out to have not as much value as they were originally thought. But nobody had ever come back to ask the question, hey, is this working the way we thought? Because they were not a level five organization. So we were able to take and reduce the intake process to about half, which meant that the agency in a single calendar here could move $1 million from administration and overhead to actually service delivery. Those are the kind of results that come from this that make this so important. And both will tell you that there's just an amazing amount of unproductivity tied up in bad data management practices. And so smoothing those out, getting them to the standards that we would like to have them, et cetera, et cetera, will absolutely deliver specific results. And we've talked a lot about states at this point, too. We've done this with a lot of private organizations as well. And you won't hear them talk about it because they told you to know their secret sauce, which is perfectly fine. Imagine how did you guys beat Amazon in the marketplace? Well, we just managed our data better than Amazon did. I'm making that up with a completely fallacious statement in there. But it does come down to that in some cases where if you just improve this, if you're doing something even a million times a day and you can improve it by a little bit, that million's going to add up pretty quickly, pretty fast. I'd like to answer Cheryl's question about selecting a DMM as a maturity model. Can everyone see the questions? Shannon? They can. Yeah. Well, some of them come in directly to either of us. OK. So anyway, Cheryl's question is that. So we should be asked in order. OK. So Cheryl said that her organization is implementing a data governance program and selected DMM as a maturity model. How long does it take to get from level one to level five? So that is essentially dependent on where you start. And I know that sounds like a trivial answer, but it's not, right? So if you have data governance implemented and it's sustained across some kind of program, such as an enterprise data warehouse or something that is multi-business line usable, then it usually doesn't take as long because your participants are more aware. So that would be sort of a level two example. If you're starting with nothing, it's going to take longer. And we would recommend that you would start with data working groups around a specific project that is important to the organization, from which you will evolve more generic structures that you can participate in and sustain over time. And level five is there as a target to shoot for. We don't really recommend organizations put in the time effort to get through level four to level five unless they have a business need to do so. So if you're in an industry with many, many competitors, then you might need to be level five in some areas of the DMM. But if you are level three around all of the process areas, you'll have a very strong program. So it's up to you to decide how much time, attention, and money you're willing to put into process improvement. And let me add a touch on that as well, too. It's not that you want to do specific global types of things. But really, this is about finding out where your pain point is and addressing it there. It's real important to let people understand that this is a program. It is not a project. So this is not something that you would expect to achieve significant results organization-wide within a calendar year. In fact, these programs typically outlast most CIOs, which has been one of the harder selling points that we've had. So remember, most CIOs are only around two to four years at this point, and anybody that starts off on this journey is unlikely to finish it. That said, but as Melanie correctly said, if you focus the efforts down into a specific subset of the overall functionalities and others, we're not going to try to do this for the entire HR department, but we're going to do it for this aspect of the HR department because it's been identified as mission-critical and therefore deserving of this additional focus of looking good for putting into it. So going back to the presentation here, it's slide 46. Is this biased? Is this based on an industry and not all on one bucket? For example, financial services as a separate dimension, insurance separate, et cetera? Melanie, do you want to address that one? Oh, this is across all industries that have had comprehensive assessments by either ourselves or our enterprise data management expert partners. We don't have really enough data at this point, to segment it the way I was describing. I had done a previous version of this earlier, which was about 500 companies in there. But you've mentioned that in one of the slides earlier that there were about 1,900 of these assessments that had been done last year. And I know you're making as much effort as you can to capture as much of that data so we can come back and say, if you're a small airplane manufacturer, here's what your data management practices look like across the industry. We'd love to get there. Most people don't really want to pay us to do that kind of research. So it sort of catches can in there. But luckily, again, we have a great team on the process. So we're doing it the 80-20 rule. We are accumulating every time and not only presenting it to the organization that we or our partners are working with, updated, but also that we're updating this industry average. And presenting it at conferences and things. This is the other part that you won't see from the other assessment, is that Melanie's always out there on the stump. If the two of us aren't together, she's in one place and I'm in another. And there's a lot of financial services organizations in that grouping, by the way. Hedge funds and insurance and mutual funds and banks, lots of them in the ranges there. So speaking of implementations, the question is, can you possibly repeat the pilot program Kaiser conducted in Oregon? And the answer is, we would love to. What we would love to do, I was just talking about that today with a health care firm who does testing for interoperability. And I was saying, they have a number of, they've been in the health care industry doing this for some time and I said, find me an organization who wants to do a big splashy pilot and a case study in a webinar. Because of course, we believe in the mission of our client, HHS, and we would love to get the PDDQ out there further. They were supposed to announce it at a huge conference last year, but they had sudden funding cuts and they weren't able to go to conferences. So it's, although it was released at the very end of 2017, it hasn't been out there with fanfare and trumpets. It is included in some fashion in some congressional legislation about interoperability, which we of course hope gets passed. And then you'll hear a lot more about it. But we'd love to do a pilot. So if you're a health care organization, give us a call, we'd be more than happy. And Milly, I'm sure you get your information in the follow-up email as well out to everyone. So what would your general recommendations to approach a data management implication in a national statistical office? Well, I would say that statistics is a great place for individuals and groups to manage data pretty darn well because the statistics function is one that everyone is depending on you for the right numbers, the right calculations, the right data verification. I can say from Federal Reserve Bank statistics that they not only have developed over 2,500 data quality rules for their main reports that they get from their 5,000 plus holding company banks, but they also do secondary and tertiary analyses and pre-calculations. So there's a great deal of effort and care taken in delivering and reporting good statistics. So I would say not knowing exactly what your organization is but I would say you'd have a good shot at scoring pretty well on this benchmark. And I'll add on to that too. Dama International gave its award to an organization and a leader in Mexico that is very tightly tied in with the government. And again, let's just go back to Melanie's example of the Federal Reserve. The Federal Reserve is looking at the economy and deciding whether to increase or decrease interest rates based on data. We kind of like them to be correct on that. We don't want too much inflation. We also don't want interest rates too high. And if they have bad data manager practices, which they don't, then we would have problems with the economy. So the economy has been running well and I think it's a testament to the practice that this is working, but back to Mexico. The folks in Mexico City are doing a phenomenal job. Their government is much more interested than the US government in these areas. And I think there's a little write up on the Dama website that you may want to look at it or find out more. We'll be happy to put you in touch with these individuals. They are just a phenomenal group as are the Federal Reserve folks there as well also. And Peter and I got to meet them in person last fall. Yeah, that was wonderful. So is there any minimum organizational maturity level required to start a DMM initiative, like have process in place or well-defined roles? You can get the processes in place and define the roles as you start developing your data management program. And as I say, as Peter says, I'm going to quote from Peter because this is his famous saying, which I totally agree with, you must walk before you can run. So if you have no data management organization and no data management program, I'm taking an extreme example, then you just want to start with one project with an eye to coming up with reusable processes, guidelines, templates, et cetera. You don't want cowboys and Indians. So if you're in an organization that is highly competitive and hostile, there's not that many of them, but let's say you're in a cowboys and Indians type of organization, data management isn't going to work for you. So you need to find some champions in the organization and you need to find some kind of collaborative effort that you can point to or hook on. For example, in the state of Arizona, the Arizona management system, because it uses lean principles and everyone involved in coming up with those process improvement suggestions, is engaged in analytical thinking. So you take the same kind of analytical thinking and the joy of that and you apply it to the data management and then you've got it done. So those are my suggestions. So Peter, I know you'll be able to talk very well to this having, being both a member of DMM and being an advocate for the DMM. So what's the difference between the DMM and the DMM philosophy and the DMBock? Well, as Melanie said, we're working to harmonize with two of them. They were independently created, but I think they were both done with the same goal in mind, that we want to support our community of data managers in doing the best job that they possibly can for their organizations. So Dana organized the DMBock wheel that we saw in the DMBock version one, and that may be where the previous question was coming from. The first version of the DMBock wheel was a little different than the second version. It's there, I could go into more detail if you want, but it's an evolution in there. So the first version went out and at about the same time, so I think that went in 2009, you have to remember all this is being done by a volunteer organization for the most part. And then Melanie's DMM came out in 2014, but I think we started talking about it back in 10, I think Melanie is when we started talking about it. So there's been coordination between the two, but nobody's ever said at Dama, we should absolutely pull this into the DMM and nobody's ever said at Isaka where Melanie works. We need to get the DMM there. We all believe this, and so we're all kind of working behind the scenes to get this to the next level on this. I don't know when that's actually going to occur, but with as much space and attention that's been going around about this whole area, I think it's a very good process. So the DMP sort of lays out what the subjects are and the DMM talks about how the subjects can be implemented in a most effective way. Yes, and we also are not prescriptive in terms of exactly what project management methodology should use or in the model, we don't tell you what are the steps to implement metadata repository, for example, when we or our partners are working with an organization, we will outline those things for them because it's a specific implementation opportunity at an organization, but not in the model. The model is the what, not the how by design. So there are some perspective differences, but essentially anything in the DIMBOK 2 is probably not going to disagree with any similar content in the DMM because all of the authors of the DMM, which was done top down, were enthusiastic users of the DIMBOK 1. So it's all the same industry, essentially. Best practices in the end are best practices, wouldn't you say, Peter? Absolutely. Can these assesses be completed remotely or are they better in person? We have in assessments used video conferencing, like the first one I ever did was for the Securities and Exchange Commission, and there were six video screens and 40 people in the room and about another 10 of them on a video screen. And that works fine. As long as they can see you, they will participate. If people are on the phone, even if they're a CIO or something, they're not going to participate. They're going to play solid there or look at their email. So video conferencing is good. You can have a lot of people remote, and usually we end up with a combination if it's a distributed organization. And in terms of being completed remotely, the other thing you're getting at, I think, is can you do this by writing or a survey or not involving human beings and discussion? And we say, of course you can use a DMM like that. You could put two people in a room, just Peter and me, and we could look at an organization and scope it and size it, but it wouldn't be accurate and it wouldn't give you what you want. Because if you're an organization, what you want is better data and better managed data. So for that, you need all the key stakeholders that are responsible for the data to have a consensus and also they also learn from each other. What are the great things that have been done with one project or one program that nobody knew about that the organization can leverage? So you want to pull things together, get efficiency and make rapid progress, you need to get the stakeholders together. Melanie, I just got a comment in here from one of our wonderful audience that comes in, this is a guy named John Toulsey, who we've loved for many years and he took my babbling that I had earlier and continuously said, I think a better way to describe it. DMM, to me, is process capability improvement. Dimbok is practitioner guide and I think that does encapsulate it very well. So thank you, Don. I look forward to seeing you next time I give you a beer. Are you handing out beers, Peter? Yeah, although I need a wine better, so. So I don't know if either of you will be able to talk to this, but what about the Enterprise Data Management Council's data management capability model versus the idea? We're always asked, here's what I would say, there's no verses in these questions. Our profession has been going on for like 40 years, right? It used to be called data administration back in the day and it has built and built and built. The governance that we have now was built from the templates of project governments and configuration control boards and governance teams to build the first data warehouses and so on. So our industry has been evolving for a long, long time and Peter, you know better than me, but I think probably if you had the stack of all the wonderful books and articles that have been written about data management over the past 40 years, it might be at least a quarter of a mile high if you pile them all on top of each other. So I consider that most frameworks are useful in one way or another. Let's take the Gartner framework, for example. That one is not a measurement instrument. It is a more of a project management approach for top down how to implement a data architecture and data management. And they talk about specific technologies which they highly advocate that you have. So that's a project management, more prescriptive approach. The DKAM, of course, was developed from the original DMM, the version in fact that was spring 2012, that the Enterprise Data Management Council, the Software Engineering Institute and the key sponsor, Booz Allen Hamilton, developed it jointly with the 50 authors. So that model is actually used now, that version of the DMM is actually used now by a number of the big banks who are participating with us. So there's a very large well-known insurance company, for example, that we keep saying, guys, you should adopt the DMM 1.0 now, it's so much better, it's more streamlined. And they say, well, we automated that first version. So anyway, the DKAM essentially was written by Mike Atkin and John Botega, specifically to narrow the scope of data management to BCBS 239 for their constituency, which is a financial group organization. So if you are worried about BCBS 239 and you want something that is easy to use and doesn't give you the kind of strategic breadth that our assessments produce, by all means use the DKAM, it's perfectly sound. And everything in the DKAM that has similar content should agree with the DMM, because we all started together. So that's the answer about that one. I love it, and just to wrap it up, earlier Melanie said that the best way to get started is to get executive by it. And I know so many people struggle with that, just to impart that data is so important and the proper management of data is so important. So really, do you have any top five or top three advice points of how to get executives to buy in? Yes, I think I can tell you what I consider to be the best and the quickest way to do it. Find a way to talk to key business executives in the different business areas of your company. Invite them to lunch or schedule a half hour meeting and ask them respectfully, but bluntly, what do they really want from their data and what are their biggest issues with their data? And if you are able to get that meeting, you'll find that they're going to give you an earful, both on the positive side and the negative side. For one insurance company, for example, we interviewed all the executives and we said, you know, if you could do what you really wanted to do with analytics, with deciding what your cash reserves need to be for any given year and property, if anything you want, what would you like to do? And we ended up with a list of 15 business opportunities that would either expand the organization's business, expand its customer base, or lower its financial risk. So if you go to the executives and you ask them aspirations and challenges, you can bubble that up and make a pitch to your CDO, your CIO, or your CEO. I'll come at it from a slightly different perspective that's also been helpful as well, which is we've talked a lot about Jeff in Arizona and Arizona certainly is on the leading edge of this and has put some tremendously impressive efforts. Isn't it sad that all the other 49 states are gonna have to do the exact same thing? Now, Mel and Jeff are committed to making sure that everybody knows what's going on in Arizona and will be sharing there, but imagine if that's across the 50 states, what's happening in your organization where you've got hundreds and hundreds of knowledge workers who are having to learn this stuff over and over again on their own and are not aware that there are best practices in the area and it's just a real, real shame to do this. So you've got to find somebody who that argument resonates with. So one of the things I do is I talk to a lot of people like Melanie does, but very often I'll find that somebody's just not interested and that's perfectly good, right? There's lots of things to be interested in and this isn't one of them, but when you do find the people that are interested and start to work with them and get them to be the beach head, if you will, and get them started on this. Well, Peter and Melanie, thank you so much again for this great presentation and content. Love having you both on as always and I hope you all can join us next month and thanks everyone for all the great engagement and questions coming in. I'm afraid that is all the time we have for today. Again, I will send a follow-up email by end of day Thursday with links to the slides and links to the recording of this session. I'll make sure that we get a contact information for you, additional information for the DMM as well from Melanie. Again, Melanie, thank you so much for joining us. Peter, thank you as always. I hope you guys both have a great day. Hope everyone has a great day. Thanks, thank you Shannon. Thank you so much for this pleasure. Thanks, oh.