 Hello and welcome. My name is Shannon Kemp and I'm the Chief Digital Manager of DataVersity. We'd like to thank you for joining this month's webinar, New Year's Resolution for Your Data Management Strategy. It is part of a new monthly series of webinars from IDERO at DataVersity. 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. For questions, we will 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 DataVersity. As always, we will send a follow-up email within two business days containing links to the slides, the recording of this session, and additional information requested throughout the webinar. Now let me introduce to you our speaker for today, Ron Huzenga. Ron is the Senior Product Manager of Enterprise Architecture and Modeling at IDERO. Ron has over 30 years of business experience and IT experience as an executive and consultant, spending a diverse range of industries. His hands-on experience in large-scale enterprise initiatives included enterprise and data architecture, business transformation, and software development. His background provides practical, real-world insights to enterprise data architecture, business architecture, and governance initiatives. And with that, I will go to the floor to Ron to get today's webinar started. Hello and welcome. Thank you, Shannon, and thank you for having me today, and hello and welcome to everyone who's attending today. Thank you for joining us, and let's kick off the New Year in style and talk about some things that we can be doing to really make a difference in terms of our data management and our data management strategy in particular in our organizations. Now, the title is a bit of a catch about New Year's resolutions, because one of the first things I'm going to talk about is let's talk about those resolutions and just see how effective they really are in general. Then what I want to talk about is how can we really implement lasting change? Obviously, this applies not only at a personal level, but also what we're talking about today is, of course, implementing lasting change in our organizations, particularly from a data management perspective. And that entails data management strategy, the data management challenges that we have to overcome to be able to implement that strategy, and also charting a course for what's our path forward and how do we actually continue to embrace the ongoing changes that we see in our organizations. With that, and obviously some final thoughts after that, with that, let's actually start talking about those resolutions. As you can probably tell, I'm not a fan of resolutions. That's why this slide is called resolutions and why they don't work. Interestingly, when you really look at it, I mean, New Year's resolutions quite often do get a bad rap, and the reason for that is they're very well intended, but often poorly executed. And when I was looking for some material for this, even though this picture off to the right-hand side is obviously a few years old, because it started like 2009 through 2012, I think it's an awesome example of what really happens to resolutions that we have and how they actually change in morph over time. So this one in particular, obviously this person started in 2009 with their black ink, and they went through several revisions over the next four years from 2009 right through 2012 with the different colors of ink. As you see on the different types of resolutions, they're changing. They're either not accomplishing their goals or they're modifying their goals to make up for the fact that they didn't accomplish them, and sometimes they just got an unexpected result and had to do something different, such as point number four that you see off to the right-hand side there, which I think is kind of the funniest one of all of them. So let's look at that now in another way. When we look at resolutions in general, 80% of resolutions fail by the first week of February. Then we wait for another year to go by, have another set of resolutions, and we repeat that same cycle the next year. What we really want to do, though, is we want to get over that. So here's the good news. Over the next few slides, I'm going to highlight a few different things that you can stop doing today, which is January the 15th. So guess what? You'll get rid of those resolutions now, so you'll be two weeks ahead of schedule for a change, so that's a positive thing in everybody's books. So why do resolutions fail? Let's talk about that. Often, it's several things. They can be way too ambitious, so they're just not attainable. Some are too ambiguous, so you don't really know if you succeed or not, and you can't really measure progress or success against those particular resolutions. Another thing that happens is when we try to do too many things at once, so if you're attempting to do major changes at once, whether it's at a personal level or something like your data management strategy, you're kind of setting yourself up for failure because you may have too much and then you might not be able to handle it, and what you ultimately end up with is anxiety and burnout rather than accomplishing the goals that you're really setting out to accomplish. So we want to rectify that. Let's look at some things in particular, which I'll call my pet peeves as it were in some of the different areas of data management in our industry in general. So here's some things that I think that we need to change right away in terms of attitudes and behaviors. The first one is data governance, and what I mean by that is there are still folks out there that think you can go out and buy a data governance solution that covers A to Z or A to Z if you're from Canada like I am, put it in and the way you go and life is good. It just doesn't work that way. You can't go out and buy the latest, shiniest governance solution, install it and have your problem solved. The right tools are important, but data governance is hard work. So what you really need to do is you kind of need to get over that and embrace yourself to start doing the hard work. You can't buy it, so stop trying to. You can buy the tools to help you with it, but you've really got to roll up your sleeves and do the work, like I said. Now if we look at that, let's look at that DIMBOK wheel over on the right-hand side. We need to be doing all of these things all the time and keep building out all the different aspects of data governance and basically working in a changing environment all the time. So we're never finished. There is no easy button. You can't just buy it, implement it and leave it. It's an ongoing process. Governance as well when we look at it has many different things. It's not just you. It's not just a tool. It's a real combination of the people, your processes in your organization, the technologies that you use, and probably the most difficult one to put in place is making sure that you have the right culture in the organization to truly implement, embrace, and move forward with data governance. Now on that light, there's some other things that happen to us as well. And I'm talking about expert opinions and reports that float around all the time. So in any industry, not just IT, but we seem to have a lot of them, there are all kinds of analyst reports or opinions or other things that kind of float around all the time. But just remember that what you're reading isn't necessarily industry-wide consensus. So always read critically when you're looking at those types of things. In particular, quite often the reports you're reading may be biased. So the one thing that concerns me, and I have seen people do this, we've got the latest report from Analyst X or Company X, we need to go out and do these things and implement them because that's what they're recommending. That's a very dangerous approach. You've really got to make sure that what you're doing is right for your organization, and you also have to be able to do several things to make sure that what you're reading even applies to your organization. The ones that bug me the most quite often are things like industry rankings because some people don't realize that a lot of the industry ranking reports out there are really a pay to play. In other words, they're based on companies that actually subscribe to that particular organization. So you've got to be very careful in terms of what you're looking at. The other thing that I want to make sure of is don't bet your company's future on these reports. Read them critically for informative purposes. And again, just because you pay a lot of money for it, that doesn't mean it's actually valuable. You've really got to make sure it really applies to your organization. There's no substitute for doing your own homework. And again, what you really need is that frame of reference of your own organization and you need to fit what you're doing based on the requirements of your organization, and that means tying it right back to your business strategy. If it makes sense, do certain things. If it doesn't, steer it the other way. The best title for this slide that I could come up with was just don't because it has a few things that I want to talk about. This is a cartoon that I used several years ago, but I think it really summed things up. And at that time, it was like, here's people around a boardroom table. They're not quite sure what to do, but the next big thing that was on the horizon at that point in time was big data. So it's like, okay, everybody's doing it. So I guess we need to do that to solve our problems. You can't just look at the next big thing ahead and take it hook, line, and sinker. What I refer to this as is a strategy known as management by InFlight Magazine. Just because it's actually in the headline doesn't mean that you necessarily have to embrace it right away or embrace it at all. Again, just because somebody else is doing it, that doesn't mean that you need to. Our parents told us that as kids, and it really applies in business as well. Again, you've got to come back and look at the things that are being talked about. Decide if they actually make sense to your organization based on your business strategies and requirements in that organization. Also, beware if you see something that's touted as a new technology that's the solution to everything. If it's touted as a replacement for all of your existing technology, run in the opposite direction as quickly as you can. I've been in this industry for, as Shannon said, over 30 years. It's actually 35 years this year, which just means I'm getting old and grumpy. But what it really means is, you know, there's a lot of experience there as well. I have never in my career seen a technology that comes in that's a replacement for everything. Technologies come in and we use them to augment what we already have in our organizations, and we may shift certain things as the new technologies shift, but we really need to figure out how to work the best of these things into our overall portfolios to take advantage of them. And remember, silver bullets apply only to werewolves, so if you're looking for that silver bullet to solve everything, it just simply does not exist. And of course, my last peeve is let's kill the buzzwords. I mean, we've used words like big data and digital transformation for several years now, but after their inception, they tend to lose meaning and erode the value over time, so I try to avoid those types of things. If you look at big data or data transformation in particular, they can mean many different things to many different people. In terms of big data, we're now at the point where data is data. You know, we have different sizes of data. We have different velocities, velocities, all those types of things. We just need to deal with data as a whole regardless of where it comes from and recognize that we're dealing with a lot of different technologies and more importantly, a lot of different values of data value as well. Also, basically digital transformation. Again, many different aspects come into play here. At this point in time, almost all organizations have embraced the digital economy and we're using digital technologies in our organization. We're at the point now where if they haven't embraced it, transformation is beyond what organizations are going to need. They're going to need resuscitation if they actually haven't gotten on the bandwagon and started to embrace these technologies already, so we're almost too far gone to start talking about that. We've really got to start talking about more specific things that we're implementing rather than these umbrella terms that are kind of used as a catch-all for everything else. So with that, how do we actually implement lasting change? First and foremost, for every specific thing we're looking at, we need to have a defined target. Unless you have a target, you don't know whether you're succeeding or not. What we need to do then is we need to take that target and we need to break it down into small, sustainable, and implementable changes. That's how you overcome that overload that I talked about in terms of trying to do too many things at once that just simply overwhelm you. Plan, then execute. When you're doing the planning, make sure you're incorporating contingencies as well. As a bit of background, I'm also a pilot, so one of the things that we say quite often is plan the flight, fly the plan, and that includes contingencies like alternate airports, based on weather reports and everything else. There's no substitution for good planning. You can't just launch and then make it up as you go. The same thing applies to business. You want to plan where you're going and you also want to make sure that you're executing against that plan, but that plan needs to be flexible enough to actually incorporate not only anticipated or possible changes, but also have enough flexibility for unanticipated changes along the way as well. Without a plan, the chance of success is virtually zero. And one of the things that has been said quite often, and it really applies here, is hope is not a strategy. If you just embark on something and think it's hope it's going to work out, you're probably not going to succeed. You really need to have a defined plan to get there. The plan needs to be concrete. It needs to be measurable, otherwise you don't know if you're succeeding or not. And something that I tout is a continuous improvement approach. And what that means is as you're progressing, continually evaluate, measure, and adjust accordingly. Rinse and repeat, and then as you're making success on certain things, then you start adding in incremental changes so that you can actually accomplish those goals. That gets rid of the big bang approach where you're really trying to do too much at once, and it just allows you to build your momentum and really start to accomplish those goals. Somebody who just came in might actually think I'm talking about agile development, because it's interesting that a lot of those same principles apply, whether we're talking about personal improvement, business improvement and refinement of strategies, or specific implementations like agile development processes. Those same principles apply across a lot of different areas. Now that we've talked about that, let's talk about data management strategy and how we're actually going to apply some of these things to data management in general. I like to use this one once in a while because it really sets a good backdrop. And obviously it's several years old now, but data management strategy is vital to our organizations. And Tom Peter said it very well in 2001 when he said, organizations that don't understand the overwhelming importance of managing data and information as tangible assets and the new economy will not survive. And we've seen lots of casualties in the marketplace because companies just do not manage their information correctly or don't understand it. To back that up, and I did talk about this in my previous webcast in November as well, and I got a lot of good feedback from folks, and I wanted to reemphasize some points here as well. And what I'm talking about here in the next couple of slides was based on a 2015 PWC Iron Mountain study, which was called Seizing the Information Advantage. When we look at resolutions and retrospect and everything like that, we really need to evaluate critically how we're doing as companies and how we're doing it as an industry. And unfortunately the answer is not very well as a whole. But when we look at studies like this, we're still seeing the same things. Very few organizations are really utilizing information to its full potential, so we have a lot of room for improvement in that area. Looking at it more generally, there's quite often deficiency in technical capability, sometimes there's a skills gap in organizations, and more importantly in a lot of organizations there's still lack of a data culture that really allows you to move forward and actually abreast data and data governance correctly. What we also see is lack of investment in what we call value-driven strategies. And what we're talking about there is strategies that are actually going to derive business value in your organization. So you need to focus on the things that are most important that are really going to contribute to the organization, like this bottom line, whether it's increasing revenue, market share, decreasing costs, those types of things are what we're talking about. The interesting thing is still very few organizations don't understand how to derive maximum value information that they have. And what that happens there is that really starts to erode corporate value if it's not corrected. Based on their study, it really is an epic fail across the industry. There are very few companies that do this successfully even in this day and age, so those are areas that we really need to start to look at and say, how do we start to improve? To back that up, I'm going to talk about the way they've categorized certain things as well. And they talked about an information management disparity. So they categorize companies into a number of different groups in this study, but one they called the misguided majority, and that was 76 of companies in the organization. And I'm not going to talk about the middle groups, but just the two extremes. The misguided majority is obviously the largest group, and there's several things going on here. Quite often organizations are informed, but they're really constrained in terms of actually taking advantage of the information for many different internal reasons. In other cases, organizations just are uninformed and just ill-equip the deal with data and understand and get value out of data at all. Some symptoms of this is, you'll quite often see this in organizations where data is seen by a byproduct, or it's taken for granted. It really has to be part of that value chain in the organization, along with the other things that you do in the organization. If you have a low comprehension of the commercial benefits that can be gained from harnessing that information, you're probably in that misguided majority group. Other things that happen is quite often organizations may be constrained by some of their legacy approaches to doing things, and they haven't modernized their approach, or they may be constrained by very binding regulations that's actually preventing them from really deriving value out of their data as well. Again, characteristically, what we often see in these types of organizations is a relatively weak analytic capability, or you might see a very strong analytic capability, but lacking value focus. What that means is you're able to do pretty good analytics, but if you're not focusing on those things that truly derive value, you're not really yielding the results that you should be able to get. Or in other organizations, there's still a very low level of analytical capacity in general. Also, in a lot of organizations, and this is getting worse all the time, it's just getting overwhelmed by data volume. There's so much out there. What do you focus on? And if you haven't properly categorized your data and that sort of thing, it just contributes to the problem, so we'll be talking about that a little bit later on in this discussion as well. A sure sign of failure is that in an organization, you still think that data is the domain of data architects. Everybody is responsible for the data in an organization, and the business owns the data. So I think it's Bob Siner that says it very well. He says basically everybody's a data steward, so get over it. Everybody has to embrace the value of data and make sure that they're contributing to data quality in the organization. And again, if you're looking at things that are IT-led rather than business-led, again, you really need a business-led, business focus on data governance in general. And again, you might see things like spreadsheet hell that's occurring in your organizations because you've got so many things going on, and you've got all these little independent silos of information rather than an overall coordinated information strategy and the analytics tools that actually pull all the information together rather than all these little independent things that are going on. Now let's look at a contrast on what we call the information elite, which were categorized as the top 4% of organizations that really got a lot of value out of their data. The causal but something proactive action, and what that meant is they were able to utilize data effectively to do very important things like diversifying the business models of the organization to take advantage of new market opportunities and everything else. Improvement of operating efficiencies, identifying and implementing new market opportunities. Looking at tangible data value, and what we mean by that is the value of data is linked directly to organizational key performance indicators. So whatever performance indicators are important to demonstrate business success, we're also tying into those when we're looking at data value and data quality and other metrics that we're tying in to make sure that we're really supporting those key organizational strategies. Being able to exploit data for competitive advantages is extremely huge and we have lots of companies out there that do this very well in terms of let's look at companies like Amazon as an example. They're a logistics company, but what do they do? They actually exploit and utilize data to really be the middle man in all these transactions that are occurring in the marketplace and they're pairing buyers and sellers and they're doing it all with data. That's a perfect example of really using data for competitive advantage. A balanced approach between security and value extraction. What that means is you have balance in terms of making sure that your data is adequately secured and only the right people that should be using the information have access to it and that balancing that and making sure you're still extracting the maximum value from that data in your organization. And again, in terms of data governance, it's not thought of as an add-on or initial project or something like that. You have to do, it's actually woven right into the fabric of the business. Governance is just a normal part of what these organizations do in their day-to-day business and that's something that you really have to be able to do. And again, what you really need to drive all of this is a well-defined information strategy and that means it's an information strategy that reflects and is tied into your business objectives in the organization. Interestingly enough, when they looked at this, they often found that the information elite had a higher concentration in the areas of healthcare, manufacturing, and engineering. When I think about that, I think the real reason for that is these three types of industries in particular often have to have a very disciplined approach in a lot of the things that they're doing in terms of record keeping and other things. So I think it just really predisposes them to really be able to take advantage of data in general. So I think if we just look at a more disciplined approach in all of our organizations, rather than what we see sometimes, which is called the Wild West approach, we'll succeed more. Let's talk about this in the context of data maturity as well, because this ties directly into it. I've used this slide before, and generally speaking, this is based on things like the capability maturity model where we typically use a scale from one to five to represent a low level of maturity, up to five, which is a high level of maturity. Something that I started doing years ago is I actually incorporated a zero on my scale as well because I have run across organizations, both when I was consulting and still in my current role, where there's almost no formalization at all. So there's an extremely low or no data maturity. And what I've got here is rather than a lot of different things, it's just some major categories that you can look at to kind of self-assessing your own organization where you might loosely fit in some of these different categories. So if you look at data governance as the first line, if you have no data governance, that means you're basically at a none. You're very low on that scale of data maturity. If you started instituting it at maybe a project level, you're getting a little bit better, but still a very low-end scale. Even a program level is getting to you to a managed approach, but you're still not where you need to be until you get to a true enterprise-wide data governance that's at the top end of the scale. As part of governance, of course, master data management. And that means how you really classify and look at your master data in the organization. The master data is extremely important because when we look at it, the master data are those things of interest to us that participate in virtually all the business transactions that we conduct. So what I'm talking about here is things like customer information. If you're a retailer or something, your product information, employees or if you're in health care, your patients and those types of things, all the things that are really important to you. So you really need to find a way to classify those. If you don't have a formal classification, again, you're down at that bottom end of zero. If your master data isn't truly integrated into other things, then you're still at the low end. And you start progressing when you start doing things like having a shared master data repository. Then when you start getting to the point where you're centralized that and you've got data management services that utilize that, down to the point where you actually have data stewardship and even a data stewardship council established in your organization, that's when you're starting to get up to the top end or optimized approach. Data integration, same type of thing. If you've got a lot of spaghetti out there with ad hoc point to point integrations, you're probably at the low end of the scale. If you're starting to move up the chain, then you're really starting to look at things like common integration platforms or middleware and that sort of thing. And then finally to the point where you've got data excellence and a very tight integrated architecture throughout your organization. And then data quality is the last one. Again, if you've got a lot of scattered silos, you're probably at the bottom end of the scale. What you really need to do is then get up to the point where you have full data quality management in your organization and that includes things basically going from A to Z. Now, one thing that you've got to remember in terms of maturity is it's a journey. So if you find yourself at a zero or one today, don't expect to see yourself at a three or four tomorrow. It really is a journey, so you're going to have to step through these different levels to get yourself there because it is a lot of hard work to get there. Also, when you look at just overall behavior and the organization as a whole in terms of these categories, if there's a lack of awareness or denial about the importance of data maturity or quality, that means you're probably at the bottom end. If your environment is characterized by chaos or it's very reactive, you're still kind of at the lower end of that scale, but when you really start to get to that stability level or you're very predictive in knowing what's going to happen with your data at the top end, that's when you know you've actually arrived as it were. Other aspects in terms of organizational focus? If you find it from an IT perspective that your primary focus is on technology and infrastructure, you're probably leaning towards the bottom end, but when you're really looking at focus on information and strategic business enablement, that is basically a formula for success, so you're probably going to be migrating more towards the top end. If you're at a low level of maturity, you have a high level of risk and a low level of value generation. If you're at a high level of data maturity, you have a lot lower risk and a lot higher level of value generation in your organization as well. So how do we start to accomplish some of this? I've talked about business alignment a little bit, so let's talk about a few different things. The top three boxes on this are all on the business side of the equation. Again, it's really about defining what your business is about and then being able to tie your data strategy and your data management to line up with what the business is trying to do. A lot of organizations may have a vision statement, which is really their large global statement about how they want to change the world. It's really got to be something compelling. It's got to be exciting. It's kind of their Sunday statement of, this is really what we want to be when we grow up, for example. A lot of companies might not have that, they might just settle on a mission statement, which is just fine. It really depends on the organization. But the mission statement is, what is it that you're really trying to accomplish? And think of this as kind of the everyday statement. What are we doing to accomplish the dreams every day? And this is often accompanied by a number of goals and objectives that support what that corporate mission is. Then we have business strategy tied into that, like I said, which is the goals and objectives. And these are really kind of the building blocks that you're going to use to implement the corporate mission or that you set out. What you need to do is you need to tie your data strategy into that. And the way you do that is you actually look at those specific business goals and objectives. And you actually set up data strategy statements that match up to those and how you're going to implement those from a data perspective or what you're going to do with your corporate data to help you fulfill those objectives. So that's doing things like tying to the, you're going to have KPIs on those goals and objectives on your business strategy. It's enabling your data to achieve those KPIs and also enabling your data to be able to measure that you have succeeded on those KPIs. And then, of course, from a data management perspective, this is really an ongoing discipline where you're really trying to deliver control and protect and enhance your data value in your organization. And this is where you start really implementing things like your plans, policies, your programs, and your ongoing data management practices support everything that's above that. Let's talk about vision versus mission just for a moment. So like I said, if it's a vision statement, it really is the dream. So how does your organization want to change the world? Someday we're going to do this. And it's typically a big, exciting, compelling type of statement that you want to do. It's really kind of, this is the dream statement if you want to put it that way. A mission statement is really what are you going to do to accomplish that dream? What is it that your company does from a business perspective? Who are you benefiting from doing it? How are you doing it? And how are you repeating this and doing this every day? Also, your mission statement should not be stated in financial terms. It should really be a motivating statement that helps people out. Peter Drucker, again, a renowned management consultant, basically says the mission statement has to express the contribution the enterprise plans to make to society, to the economy, or to the customer. And basically those that express their mission statement in terms of financial terms usually fail to create the cohesion, dedication, and the vision of the people who have to do the work as to realize the enterprise's goal. So what that's really saying in a nutshell is if you're an employee of an organization and your organization has a mission statement, the asset test is whether that vision, that mission statement is actually motivating you as a member of that organization. And again, business strategy is extremely important. Everything that we're doing in IT or data really needs to support the overall business strategy. The business is our reason for being. We're not here for data sake. We're here to actually support and make our businesses successful. So what that means is we also need to have a supporting goals and objectives both at a business strategy level and at a data management strategy level. And those goals and objectives need to be quantifiable and measurable. So what do I mean by that? That means that we need to make sure that we're utilizing, using what I call smart metrics in our organization. And if I break down the word smart, it basically means the following points. Number one is the metrics that you're using need to be specific and they need to be properly target the specific area that you're measuring. The next part of it is you need to be able to measure them. So that means you need to be able to collect data that's accurate and complete to tell you, you know, how you're actually attaining those goals. They also need to be actionable. And what that means is you need metrics that are easy to understand. It needs to be clear when you look at your performance over time, you know, whether you're actually achieving success or you're not. And so that you know where to take action. So that may be building in certain things like thresholds and that type of thing where if you see values shift, you know, outside of a certain range, that that might be a call to action to take corrective courses as an example. And again, the metrics need to be relevant. And again, start small. Don't try to do too much. So you want to measure things that are relevant to the area that you're looking at. Don't try to measure everything and then tie it together because you'll just be introducing a lot of noise into the equation. So focus on specific KPIs that are relevant to that area and then grow out from there once you've basically handled that. It's that same philosophy about start small and then grow again. And of course, timely. We need to be very careful when we're looking at the data that we're using for these metrics as well because data can be stale as well. So the data that we're using to measure success needs to be as close to real time as possible because quite often if data is received too late, it might not even be actionable anymore. It may be too late to correct course. So let's look at some data strategy objectives as well that tie in here. And this is really, again, when we look at those information elite that we were talking about previously, these are some of the types of things that we see happening in those organizations. Again, information governance oversight is comprised of all key functional areas of the business. It's supported by the senior leadership in the organization. And again, governance is owned by the business. It's not owned by IT. What's also there is a real culture that is evidence-based decision-making. In other words, you're tracking information. You're having metrics. You're basing your decisions on the results of that measurement and the metrics that you're tracking, rather than on conjecture or hyperbole. And information is really perceived as a valuable asset in those organizations. What you also need to do, of course, is we live in times where, of course, data is extremely sensitive and can be misused and everything else. So all of our sensitive and valuable information needs to be adequately protected and secured. And the secure access is granted only to those that actually need it. And again, fit-for-purpose data analysis, interpretation, and visualization. Fit-for-purpose is key in data quality. If you're not using the correct data for what you're trying to examine, you could be led to wrong conclusions. So you really need to make sure that the data you're using is actually appropriate to the decisions you're trying to make. And again, underneath all of this, I can't overemphasize this. The way you accomplish this is sound underpinnings of data architecture and enterprise architecture in general. Things like data modeling, which really help you to understand the data in context and how it relates to the other types of data in your organization. And the business process modeling, which really helps you to understand visually how the data is created and used in your organizations. Without these visual aids and visual models, it's extremely difficult to understand data and how it's used in your organization. Again, and you've seen this slide for me before if you've been in one of my webcasts, is one of the way we look at it is you need to have that overall structure, and you need to be grounded in that solid foundation of data architecture, which is then supporting the other pillars of enterprise architecture such as your business architecture, is that central pillar, your application architecture, and your technical architecture. That's what's driving your enterprise enablement overall, and that supporting structure helps you to enable governance and by tying all these things together from a data perspective is how you're actually able to start to accomplish those strategic goals and objectives from a business perspective as well. So let's talk about some specifics now that we see in our organizations in general. I've actually pulled out an old slide here. It kind of came up in a discussion that I was having in an enterprise architecture group last week, and I thought, you know, it actually made perfect sense to pull this slide out again for this particular presentation as well. A lot of organizations are talking about data lakes, and they're trying to embrace data lakes, and frankly, they're struggling to do so. So again, our data lakes can be comprised of a number of different things, everything from data out of legacy databases to location data, stream data that's coming in from social media and other areas, click streams from websites, and also production metrics and other things. So there's a lot of stuff that may be floating around in what we think of as a data lake. Many years ago, I talked about the concept of the data swamp, and I think that we're seeing that more and more now than we have in times gone by. A lot of organizations are trying to embrace the technology, but they're not able to handle it well. So what they've really got, it has ended up with this dumping ground where they've got a whole bunch of data in these data lakes, but they don't really know how to work with it or get value out of it. This next slide has double meaning. Quite often, and for years, we've talked about our organizations as information refineries. It's basically a phrase that Bill Imman coined many years ago, and he really talked about the disciplined approach of actually being able to extract the value out of the data by making sure that we are working with our data correctly. Obviously, it was tied very much into the idea of BI and that sort of thing. But again, this slide has double meaning. If you go back to that previous data lake slide where it's been kind of dumping all the data in and not really knowing what to do with it, rather than information refinery, this very quickly turns into a sewage treatment plant because what you really need to do is you need to be able to clean all of that data before you even have a fighting chance to do anything with it at all. So you've really got to make sure that you're handling that data correctly. This is what I call one of the major challenges that we're up against in organizations. I used to call things mapping out the data landscape and those types of things, but what we're really dealing with now is what I call the multi-hybrid data ecosystem. And the reason I refer to it now as an ecosystem is we're dealing with increasing volumes of data, a number of different technologies that are proliferating out there, and we're seeing this stuff also grow organically in shift on almost a daily basis in terms of the types of information that we're dealing with. So rather than just being kind of a static thing that we're looking at, it is really something that's truly evolving, so we need to take more of that biological or ecosystem type of approach in terms of trying to understand and address it. So I'm going to start this in terms of some builds from left to right, just to kind of characterize it in terms of we have a lot of different types of data feeds and things like that that we're dealing with. It could be things that are coming out of no SQL sources. Some of it comes out of our relational databases. We might have streaming data, either from sensors or those types of things, social media feeds, flat files, spreadsheets, and other data feeds. All this type of information is kind of coming into our organization or produced in our organization. So with our technologies, we really need to ingest that information and we need a place to land it. So what we end up with is what we call the raw transient data. Then quite often we find this thing, this thrown into things like Hadoop clusters or those types of things, because the technology enables us to quickly get the information in there, because it's things like Scheme on Read and that type of thing in some of the technologies. But we also have these sandboxes because we realize we're onboarding some of this data that we don't know what we're doing with it yet. So we need to do some experimentation and profiling and that type of thing in those sandboxes from a data science perspective before we even know if it actually has some value to our organization. From there we actually build further. Once we have gone through that, that's kind of an evaluation step as it were. So we get to a point where, okay, we've got this raw data. We need to get a point where it's kind of approved. So it's like, okay, we can go ahead and continue to work with that, but we need to discard the stuff that really isn't good for us. So we kind of have that stage that we have to go through. From there, we finally get to a point where we know which pieces of data we would consider our trusted data. And obviously there's a lot of discipline to these three stages to make sure that we do that. This trusted data is what we need to be using to make our business decisions. So once we've gotten to a point where we've got this trusted data, which is spread across a lot of different data stores, it could be MDM stores, it can be basically a number of different relational technologies that we may have in our organization, no SQL technologies, it's all there. So we really start thinking of this overall system as a logical data lake and we're really dealing with multiple levels and multiple technologies all simultaneously. From there, this is where that refining concept comes in. So that's where we're starting to do things like ETL, whether it's a formalized BI approach or still utilizing data lake architectures, we still need to do extra transform and load and that type of thing to understand the data. And ultimately, of course, we get to the point of using the refined data that's driving the decision-making in our organization. And the end goal, of course, is not only in our transactional systems to be using this data, but also to be able to support things like self-serve, analytics, reporting and those types of things because this is typically where we're going to be reporting on our business level KPIs. There are a lot of things that have to line up to make this happen. The way to do it is to map that data and the way to map that data is through the use of data models. So again, everything from conceptual, logical, physical models, dimensional models for BI, and also the enterprise models extremely important as a focal point that really identifies the types of data that's important to you in your organization. So by overlaying the data models to understand all these technologies, plus things like visual data lineage models to show that the way the data is actually transforming and moving through the organization is extremely important. Whoops. On top of that, hang on here, on top of that, we need to have enterprise data dictionaries naming standards to understand what we're talking about, something I call attachments, which is really metadata extensions as it were to categorize and catalog all of this data even at the model level and flow through the organization. And then from there, we have our metadata repositories which contain our business glossaries and everything else that's extremely important to us. When we come back to that data maturity slide that I talked about earlier, when we look at the use of data models in particular, each of these dots represents the scale from one through five on a data maturity level. So at a low level of maturity, you may just have to be using data models for documentation, and it may be very project focused. Whereas at the top level where you've achieved full data maturity, what you've really got is fully integrated data modeling, glossaries, metadata, and self-serve analytics. Not only on the data in the organization itself, but from a data architecture's perspective, analytics on your metadata as well. Obviously, there's a lot that we need to do from a governance consideration. Again, coming back to that thing that was on that previous slide, we need to be looking at all of the different aspects of data governance. But in particular, of course, data modeling, design, and data architecture are fundamental components of overall governance in our organization. We need to focus on those types of things. From a data classification point of view, we need to do things like classify our master data between master reference and transactional data at a minimum. We need to prioritize which data is more important to us from a business value perspective in our organization, and we need to divide and conquer approach to just really go after the things that are important to the organization itself. What we also need is we need to be able to assess data quality. So the metadata that we create should also have associated data quality characteristics. And again, we need to identify those that are critical data elements that are vital to our organization. And of course, we can never forget about the regulations and security and privacy, which is a huge part of governance and tie that information together as well. Some questions we can answer through the modeling is understanding the organizational data. What's important? Where's the data? Because it can be many places simultaneously. Where did it come from? How is it actually used? And this is how we tied into business processes with business processes as well. And what's the chain of custody? Visual data lineages is an example. And what are the business rules around that data? The data models themselves encapsulate a lot of the business rules just because of the associations between the different types of data that we have. And from governance, how to identify private information? How long should I retain the data? Again, how do I classify that master data? And is the data fit for purpose? And when you're making data changes, what changed and why? This is all part of the data modeling and the use of your data models that has to go on in the organization. In terms of the master, in terms of the metadata tags, these are the types of things I'm talking about, just as a very simple example where I've got a very skinning down entity here that just has a couple of attributes on it to make a little room here. But I've got these categories for all these different things, master data, business values. I've also got basically governance information such as privacy levels, security impacts. I drive all of that from my data models all the way through the rest of my data architecture. Again, we need to tie that into business glossaries and terms. Why? Because we need to agreed upon definitions and terms or a vocabulary that we use in our business so everybody using those terms is talking about the same thing. The way we accomplish that is through business glossaries and then we tie those business glossaries back to the important pieces of information in our organization that they define. We also need again from a regulatory compliance and just overall governance perspective, we need to identify what our policies are. For example, this one's some GDPR data policies that I'm showing here. We need to break that down into policy statements and then again we need to be able to tie those things back into specific data objects that they apply to. Here's an example in one of my repository of here's some general information about customer data out of one of my systems. I have the tied in related business terms, I have related business policies in there and I also have warnings on it because that came directly from things like security and the governance properties to tell me that the data is sensitive and it also has a very high security impact. From my perspective, stand-alone metadata repositories don't make the cut which you really quite often you've got metadata repositories out there, you're doing a metadata import from different areas or discovery to try to load the information in, then you end up with a metadata catalog and you don't have visual models to give you the clarity or the picture of what's going on. So quite often you're relegated to things like tech search and look up and really trying to manage and match things to get value out of that. I call that the Flat Earth Society because it only gets you part way there. You recognize that you're part of the planet but you don't really know how to circumnavigate that planet to make sure that you can tie all that information together. And I had to add this little one to the slide as well. I stumbled across this on a social media feed at one time and it's very interesting just think about that statement that the Flat Earth Society has members all around the globe and it'll sink in if it hasn't already. What I propose is fully integrated metadata in models and this is the approach that we utilize where we have that global perspective and focal point where we're really relying on things like the data models, the business process models, the visual data lineage, tying all that together basically pictures and words tying it all together to understand your overall metadata. With that you can then time your policies, your reference data and everything else to give you that global perspective of what's really happening with your data in the organization and how you can actually tie that together to dry value. Some final thoughts just to kind of tie this thing together because we have covered a lot. Again, abandon those annual resolutions with a continuous lasting change both personally and in your organization. Stop the bad habits, in other words no management by in-flight magazine, really read things critically, evaluate things critically and fit for purpose for your organization. Again, governance, data management, data strategy, it's all hard work. So basically roll up your sleeves and then reap the rewards of that honest hard work that you're going to have to do to accomplish it. Very important is the data strategy has to be aligned with the corporate vision mission and goals that basically guarantee success of the organization. Conduct a data maturity assessment if you're not sure where you're at and this can be a qualitative maturity assessment that maybe you do internally and maybe even use some of those guidelines that I gave you in that one slide. Or you can also go out and you can get formal data maturity assessments from organizations or self-assessments that you can get as well. The important thing is be realistic and honest about your starting point because that's going to help you plot your strategy to get your ultimate goal. Smart metrics, again, make sure that you're using truly measurable metrics that are understandable and apply to the area that you're looking at. You need the metrics to assess progress. And again, make sure your data management has that solid foundation so you've got enterprise architecture and modeling. Data modeling in particular is extremely important. Remember that your models are your maps for the journey. If you're going from point A to point B would you rather do it by a set of written directions or would you rather use something like Google Maps to plot your course there? Using models is the same type of analogy there. And again, metadata repositories without integrated modeling to really tie into those visual models to help you understand just don't help you drive the value that you need to. Using the models is extremely important. And again don't take on too much at once. Start small and grow. So pilot projects to demonstrate some value if you're just getting your feet wet in this type of thing. Focus on the business areas that will drive the best returns. That basically ties in with the most important KPIs to your organization. Grow from there. Then start folding more in. Don't forget to celebrate success. It's hard work. So you need to basically help yourself. And rinse and repeat. Just keep building on it and kind of keep repeating that cycle and a continuous improvement philosophy. And again, on the personal side, since we were kind of talking about resolutions, one thing I recommend is this is hard work. Focus. So to retain clarity, make sure you have a good hobby or interest or several that have absolutely nothing to do with your work. Being able to focus on those refreshes your perspective when you come back to this thing. And plus it makes you a more interesting person in general, in other discussion and everything else. So with that, that's all I really have to say for today. So now we'll open up to Q&A. Ron, thank you so much. Oh, Ron, thank you so much for this great presentation as always and for kicking off that brand new year with this great topic. Just a reminder to all the attendees, I'll be sending a follow-up email by end of day Thursday with links to the slides and links to the recording of this session and anything else requested throughout. And if you have any questions, feel free to submit them in the Q&A section in the bottom right-hand corner of your screen. So dive right in here, Ron. You talked a little bit about this, but could you provide some examples of how to support a data management strategy? Well, data government, I mean, it's all tied together. So an overall data management strategy has to incorporate many different things. You have to look at the technology and your architecture. You also need to have the overall governance. So the real focus on things like data quality, master data management, all of those types of things, they're really folded together. You can't really do one in isolation without doing the other ones. You end up with what's an unbalanced approach and you just won't achieve the results that you're looking for. Does that kind of make sense or help? It does, I think. It does, I think. And definitely, so moving on here, so where have you included metadata management which includes classification either taxonomy, ontology, et cetera? The metadata management the way I look at it is people think of it just existing in a metadata repository and that's what I'm really trying to drive home here on certain things is it's not just something that kind of lives and resides in a metadata repository. If you start right at the outset, if you think of your data models as an example as that visual model or the blueprint of how that data functions in your business, it's not just for capturing things like the attributes and those types of things. That's where you want to start defining and building out those classifications and then that flows into your metadata repository. You can actually build more on top of it there as well but by doing it right at the modeling level it allows you to do things like expose it directly on data models which means you have visual diagrams that people can understand in reference when they see it. So for example something I do is I'll publish data models where for certain audiences I'll have a view of the model that basically has the entity and the definition of what that entity is and what it means to the business associated with a number of these different metadata categories on it whereas in other views I'll have the more technical views that show things like the attributes the indexes and those types of things so it really allows you to isolate the perspective for the audience that's consuming that particular type of information but you get the benefits of the visual associations between them as well. And the questioner was also commenting that it refers to a maturity assessment slide if you could tie it into that as well. Okay basically in terms of overall maturity to achieve a high level of data maturity that means you need to know the data in your organization inside you need to understand it you need to have it fully categorized it needs to be easily accessible so that you know where the information is to support the decisions you're making and that it's fit for purpose and all the different aspects whether it's you know which is the models the business glossaries that are adding more meaning to all the different data assets the regulatory policies that tie in all the data quality attributes that you're tracking in terms of also those that are critical data elements which means those that are the very most important of the business. You need to know all of that so that you're using the right information to drive value out of your organization once you've done that you've actually achieved a very high level of data maturity because in order to do that you're going to have to have stewardship and everything in place that's just a matter of everyday business it's not something you think about separately it's just what went into the fabric of your business at that point in time does that make sense? I believe so we'll see if that makes sense for the question or they can certainly add more building quality into your product on an assembly line right all the different facets that go in build quality into the product on its way on its journey through the assembly line and think of data the same way sure and the question you know but that didn't seem to be a criteria in your maturity assessment but I think you certainly so if you want to I give some fairly I guess qualitative or broad symptoms there that kind of are indicators of when you might find yourself in one of the different levels or categories so just using that in a sort of to help you guide yourself through a qualitative assessment will help Alrighty Ron there seems to be confusion here on the ER Studio products yeah if you have a quick overview of the different modeling tools there sure so basically I guess is a 30 second commercial as that our point of clarification ER Studio is a well fled enterprise architecture modeling suite obviously people often think about the ER Studio data modelers the cornerstone of the suite because it's been there the longest that incorporates full modeling data modeling and visual data lineage modeling in it as well the business architect product is where typically business analysts and business architects would live so they can build out high conceptual models which of course can be shared or brought over to the the data modeling tool for further elaboration but very importantly there are full BPMN2 compliant business process models as well the data stores from the business data models can be referenced in the business process model so you can get down to a very detailed level of representation in terms of tying together your business process and tasks with the data that you're working on right down to the level of specifying CRUD in terms of create read update and delete on individual business data elements if you wish to as well and kind of surround also there's a software architect which does many different types of things like UML modeling component modeling those types of things tied all together there's a repository for checking in and checking out models and workspaces and then there's also the metadata repository that extends that or what we call team server which is where we also tie in things like the business glossaries social media collaboration or social collaboration to have discussion threads and everything else and really tie together all those business and data artifacts as well as the definitions from the business glossaries and the policies that regulate them love it it's a perfect question awesome I love it and it does bring us right to the top of the hour when I thank everybody for being so engaged in everything we do and the great conversation going on so far but it is all the time we have Ron again thanks for this great presentation just a reminder I will send a follow-up email by end of day Thursday for this particular webinar with links to the slides and links to the recording and then as I mentioned it's a brand new series that we're doing with Idira the next webinar will be on February 5th covering data architecture the foundation for enterprise architecture and governance that should be a great one I'm excited so Ron thank you again so much for another fantastic presentation and I hope everybody has a great day great thanks Shannon and thank you everyone thanks all