 Hello. Welcome. My name is Shannon Kemp and I'm the Chief Digital Officer for Data Diversity. We'd like to thank you for joining today's Data Diversity Webinar, Data Management Best Practices, sponsored today by Precisely and Satori. It is the latest installment in the monthly series called Data Ed Online with Dr. Peter Akin. 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 by the Q&A section. Or if you'd like to tweet, we encourage you to share highlights. We encourage you to share highlights. If you'd like to chat with us or with each other, we certainly encourage you to do so. And open an access side of the Q&A or the chat panel so you can find those icons in the bottom middle of your screen for those features. And just to note, the Zoom chat defaults to send to just the panelists, but you may absolutely change that to network with everyone. To answer the most commonly asked questions, just 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 will likewise send a link to the recording of this session, as well as any additional information requested throughout the webinar. Now, let me turn it over to Nick for a brief word from our first sponsor, Precisely Nick. Hello and welcome. Hello. Can you hear me okay? You sound good. Thank you. Thank you, Shannon. My name is Nicholas Sather. I'm the Product Marketing Manager here at Precisely. I appreciate everyone for joining us for this Data Diversity event. I'm going to take a few minutes to talk about where Precisely sees data strategy at and how data integrity relates to your data management practices. I first want to start off with what we're kind of currently seeing in the environment around us here. No one can really deny that the last couple of years have been really an absolute whirlwind. First and foremost, everywhere you've heard it a thousand times, we all know the pandemic has really long-term impacts on virtually every part of daily life, every person in the business, and most of their business functions. Compliance with changing regulatory and privacy requirements imposed by states, industries, partners, and customers have also really forced organizations to double-check all their processes. And recent geopolitical instability has really created some economic uncertainty that requires companies to be very intentional with the resources and their investments. We also see the recent economic downturn impacting organizations and influencing their data strategy. Now, I, like you, see a lot of numbers here, so let me really highlight the key takeaways. In a recent study by Drexel University, 450 data professionals representing companies across all industries, countries, and sizes reported how they're impacted from their recent economic downturn over the last few years. 40% experiencing reduction in staffing and 37% reporting a decrease in their budgets. But the survey also showed what steps companies are taking in their data strategy to manage through these constraints. And what we're really taking away from the highest trends here on the right side is they're focused on increasing productivity and dropping down costs and resource requirements. 57% are moving workloads to the cloud and 43% are going through a digital transformation. And in that same survey, over three-quarters have responded to resoundingly reported that data-driven decision-making is still the most important goal for their organization's data programs. In order, overall, what we're seeing is indicated that organizations are being very prescriptive with their data strategies to make sure that they're seeing its value. It's more important than ever that they see an ROI on the investments in data. So in order to be very prescriptive and make sure companies can see the return on their data programs, we advocate that data strategy takes up more of a business-first approach. You might be thinking, well, what does that, what does a business-first approach really look like? The beginning step to taking an approach like this is to really prioritize the data that matters and strip away any data that isn't critical. It's one of the primary reasons programs fail is that they can't really gather and prioritize their critical data. Another common mistake is loosely defined business problems and goals. So after sitting down the most critical data assets, say, for example, in your governance program, linking them to your overarching business goals so that you can clearly measure their impact. Many times, programs aren't flexible enough for all levels to engage and take ownership. So taking a business-first approach helps you build stakeholder engagement across three levels in the organization. And then lastly, embedding that strategy in everyday business tasks. So it's not just a do once and forget it for a year kind of thing, but continuously involved in business processes. So it's clear how governance or quality programs help accomplish the business goals. But your approach to data strategy is really only half a battle because remember the whole point of taking a business-first approach is so that you can be more data-driven in this economic environment. And all that work really goes out the window if your team doesn't trust your data. This is probably an example of one of the things Peter might mention later on, but bad data plus awesome things equals bad results. So to achieve your data strategy goals with data you can trust, we believe companies need what we call data integrity. Data integrity is data with maximum accuracy, consistency, and context for confident business decisions. And there's really two things in note here about data integrity. The first is that it isn't that you have it or you don't type a binary goal, it's similar to kind of maintaining your personal health. You don't just stop after exercising once you hit like a wake goal. You continue to maintain it as you grow. And the second thing about data integrity, one thing that we hear loud and clear when we talk to our customers is that everyone's path is unique. As a result, each company will tackle data integrity through distinct projects that deliver business value to them now, addressing these current macroeconomic trends that I mentioned, and then focus on building their data integrity as they expand, but all pass point towards continuously improving the integrity of your data, better understanding your business, and ultimately better serving your customers. You may have already already have the resources to get there on your own, but if the business first approach sounds like I deal to you and you're looking to stand up quickly, precisely also has strategic services, a team that's a serious differentiator for us. And that's because they bring over 50 plus years of combined experience solving these specific business challenges. They truly live and breathe this stuff and really bring a wealth of expertise that specifically make sure that your investments, your initiatives are successful. One of the things I personally love about them is how they're designed to lean on the products you already have. But if there are elements of the equation that are needed, precisely has the data integrity suite, an interoperable suite of services that is modular. The suite contains everything you need to deliver accurate consistent and contextual data to the business. And the incredible thing about it about the suite is that doesn't matter where your path begins, whether you're trying to link data to governance rules and policies that sync to your business first data strategy or standardized data in the cloud. So your team can trust in or enrich your address data and data quality pipelines to expand upon its value. Now you can confidently execute on that business first strategy, trust your data and not just adapt to these macro economic trends, but thrive in them. And in case you haven't heard of us before, precisely offer software solutions, strategic services and data enrichment products to help over 12,000 customers deliver data that is accurate, consistent and contextual. So they have a confidence in their business decisions and their data management. Because we're the leader in data integrity, I'm sure you recognize a lot of companies here. We work with 99 of the top Fortune 100 companies and really selectively partner with the strongest data leaders in the world. So I thank you for listening. If you'd like to continue the conversation to see how what this might look like for your company or better understand how we can help. I encourage you to visit our website, reach out to set up a meeting. But I appreciate your attention. I'll pass it back over to Shannon. Thank you so much. And thanks to Precisely for sponsoring today's webinar, helping to make these webinars happen. And if you have any questions for Nick, you'll likewise be joining us for the Q&A portion at the end of the webinar today. Now, let me turn it over to Guy for a brief word from our second sponsor, Satori. Guy, hello and welcome. Hello, hello. Good afternoon. Good morning, whoever you are in the world. My name is Guy Rez. First, let me say good job, Nick. If you haven't checked out their software, I watched some YouTube videos before this session. They do some really cool stuff. And I know we're all very excited to hear what Peter has to say. So I promise I won't take too long, but just wanted to give you guys a quick introduction to Satori and maybe start with why we even exist. So Satori was founded a couple of years ago by a data engineer who was a little frustrated by the fact that they couldn't use their data. He had all the modern tools, but simply couldn't scale that data usage. There were just too many compliance, constraints, security teams, auditors kind of harassing them to implement different things. And that got to a point where they couldn't do the job he was hired to do, which was deliver data to his end users. And so what they came up with between co-founders is a system where you can define and secure data privacy and governance all in one location, giving data to the users almost magically, meaning the right level of access to the right user, salarating time to value, and that sort of thing. Well, that problem that he encountered kind of gets solved with the Satori approach. In short, Satori is a data security platform that is a personalized user data experience. Think about giving each one of your users secure time limited access to data that they need when they need it. You simply focus on creating beautiful revenue generating data products and cut down on your maintenance for innovation. Satori works by creating the singular access path that all data users take. This allows for a central point of control for implementation of security and compliance policies, essentially no more making changes to each data store to support different use cases. Think about this as the true data mesh, right? Data is typically stored where it's supposed to be stored. And the Satori system gives you the ability to create a policy for the user based on their permission, as well as the data they are trying to access. So what does this look like in practical terms? So we've been speaking to a bunch of data teams across the globe and kind of like our co-founder of finding that what's holding them back from doing anything is actually internal operations. Modern data platforms really allow us to store and analyze data at exponential scale. But we aren't granting our users access to this data. What's the point in all of this? Storing the data is kind of pointless unless we use it. And so to add to this, it's kind of rare to see single data storage locations these days, multi-cloud, hybrid multi-cloud, combinations of self-hosted, multi-hosted, and cloud-hosted data lakes. And the continuously growing list of tools and clients utilize to access these systems. The slowdown in data products is almost always an internal policy. It's not necessarily a bad thing, right? We want to implement security. We want policy to take place. But if policy is slowing down our time to data, it's also slowing down our time to revenue. With Satori in place, we reduce this time it takes to implement these policies, whether it's needing to redact sensitive information such as PII or PCI from end users. Do we need to identify which tables have health care information so only HR can access them? Or do you need to limit which rows are returned to a business analyst because of a right to be deleted or a data protection act? Most of these in and of themselves aren't very difficult, but combine these with requirements with the ever-evolving landscape of back-end data stores, and now you'll find that there's probably more data being granted to end users than they really need because we're trying to speed up time to those data products. And so what we're really asking is, Satori, what's it used for? It's used to allow data teams to discover, define, and deliver data products with turnaround times that just aren't realistic with manual implementation. What used to be measured in weeks and months can now be defined in hours and days across all your data clients, across all your data stores, across all your data products. Think about a single point of control that allows you to manifest a data idea, implement a data security policy, audit its use, and give your users time to data. Instead of sitting in an IT ticket for weeks, a self-service access to data with preemptive security means everyone's happy and everyone can kind of get to their job. And like I said, not going to take too much more of your time. Love to hear from Peter and get this thing going. Thanks, Janine. Thank you guys so much, and thanks to Satori for also sponsoring today's webinar and helping make these webinars happen. And if you have any questions for Guy, he will likewise be joining us for the Q&A portion at the end of the webinar as well. And let me introduce to you our speaker for the webinar series, Dr. Peter Aikin. Peter is an acknowledged data management authority and associate professor at Virginia Commonwealth University, president of Damon International and associate director of the MIT International Society of Chief Data Officers. For more than 35 years, Peter has learned from working with hundreds of data management practices in 30 countries, including some of the world's most important. Among his 12 books are many first starting before Google and before data was big and before data science. Peter has founded several organizations that have helped more than 200 organizations leverage data specific savings, which have been measured at more than 1.5 billion US dollars. His latest endeavor is anything awesome. And with that, let me turn everything over to Peter to get his presentation started. Hello and welcome. And welcome to everybody else, Shannon. Thank you very much. And Nick and Guy, thank you for some excellent intros on here. I'm kind of honored to have you guys set us up for this. Our topic today is data management best practices, or I like to read you the titles, Practicing Data Management Better, because that's really what it turns out to be is, in fact, a practice. Doing data better means that you're understanding the vastness and quality that plays really an increasing role in everybody's life. And you're motivated to increase your individual data skills because you know that poor data skills cost your organization more, steal increasing amounts of your organizational knowledge workers time, deliver less and present greater risks all the way around with that process. Data management is a critical skill that we somehow have not yet managed to get the universities and colleges of which I'm a part of the system to recognize it yet. And it's important to develop defensive skills as well in that process. You know, in some senses, organizations are now trying to assign increasing values to the data that you use in the organizations in order to do this. And as Shannon mentioned, my most recent topic on this is on data literacy, where I've just come back from a quick tour of China where they are really, really interested in that topic as well. And I think we need to pay attention to it every country as a matter of national priority around this. So let's talk about the program today. What we're going to do in the next 45 minutes is dive into and understand there's some frustration because most people are unsatisfied with the current state. And quite frankly, as a general category, we are not making progress. There are lots of individual exceptions, of course. The real question that comes back is why are we not making progress? And the answer is it's really a lack of a holistic approach. So I'm going to talk about the origins of what we're doing, building on some proven research that started out when I was at the DoD. There was something called the Software Engineering Institute, the MITRE Corporation and CMMI in order to look at some of these things. Industry has a push for best practices, and that's actually kind of good. In fact, it's against the law not to go use best practices if you happen to be part of the US federal government. So what are the ingredients of this? Well, there's a data maturity model and a book of knowledge. And the key is how to understand and apply them together in this because this whole data management practice is governed by something called the weak link in the chain architecture. We'll touch a little bit on strategy and the importance of a three-legged stool, which no surprise to all of you, is people, process, and technology combined and sort of finish up around the idea of how does one get better at this? And the answer is, of course, how does one get to Carnegie Hall? The answer is practice, practice, practice. We'll look a little bit at where to go next. And then, as Shannon said, invite Nick and Guy back in here at the top of the hour to do some Q&A and hopefully some very interesting topics that you all come up with. And of course, I always learn when they're on the line here as well. Let's just dive in. Measures of unproductivity in your organization is a really good place to start playing and looking at what we're talking about when we talk about improving data practices in your organization. The first thing to understand is that knowledge workers are under increasing amounts of stress. 33% of all knowledge worker time is spent reworking or recreating knowledge that already exists at only 10% of their time is spent creating new knowledge and new content around that. So that is sort of a challenge that we're looking at. But more importantly, 53% of your knowledge workers would rather do household chores or pay bills than use your content management systems and repositories that are there. 74% of them feel overwhelmed or unhappy when working with data. 33% of them spend at least one hour a week procrastinating over data-related tasks. A couple more measurements here. Everybody has a... Sorry, when we survey everybody, by the way, these numbers are coming from the data literacy project down here. We've done a very fine job of surveying these attitudes worldwide. 14% of them have a very good understanding of how to use data. And of our young generation that's coming along to take our place sooner or later, only one in five of them consider themselves data literate. More importantly, when I'm sitting around with business decision makers, 24% of them feel that they are able to... They're confident in their ability to read, work with, argue, and analyze with data. That's our very basic definition of data literacy. 33% are able to create measurable value from data. 27% say that their analytics projects produce actionable insights. And most importantly, when I'm sitting around a table with a group of executives, I can look them in the eye and say, four out of five of you would pay me under the table to make you more literate around this particular process. So let's stop pretending and actually dive in and say, we need to do some more education around this. Many organizations feel that the idea of putting better quality data in the hands of employees actually works. And you can see here again, same statistics, same surveys. When asked to do this, 52% of knowledge workers say that they will incorporate data into their activities, but 48% of them will defer to their gut. And that's a very scary statistic. Why bother doing data at all if they're going to do that? Well, that's a piece we need to work on. And it gets more as we go up the chain. Two thirds of executives will defer to their gut as opposed to incorporating data into their decisions. The lack of data skills is really limiting our workplace productivity. Again, 50% of them incorporate data. 36% say they'll find an alternative method to complete the task without using data. I guess that involves a Ouija board perhaps, or 14% will avoid the task entirely. Well, again, these are not really good nose to come out of this. Another question is, well, why haven't my data problems been solved when we moved to the cloud or built a data warehouse or invested in technology or hired a CDO or purchased Salesforce? All of these are good steps, but you've got to do them in the right sequence. Otherwise, it will be problematic. Let's just start off with sort of a basic, which is we're doing lots of data consolidation these days and data that's in a cloud or a warehouse should be cleaner. It should have these three attributes that data outside of the cloud or the warehouse doesn't have. It should be smaller in volume and it should be more shareable by definition. Let me take just a minute and explain how this works. Again, the original example here was warehousing, but it turns out it's exactly the same economics that go into the cloud because most people take their data and they move it into the cloud. You'll notice the original cloud is much larger now. And the problem with worklifting your data in this fashion is that there's no basis for making the decisions around this. There's no inclusion of architecture and engineering concepts and no idea that these concepts are missing from the process in the first place. And this is even scarier when you consider that 80% of your organizational data is wrought. Wrought is an acronym that stands for data that is redundant, obsolete, or trivial. So how should it be done? Well, again, the idea is to take and look at this point of inflection where we're taking data from outside the cloud or outside the warehouse and moving it into it and transform. In the warehousing environment, this was called ETL and transform was one of the key components there. Again, data in the cloud or in the warehouse should be less in volume. It should be cleaner and it should be more shareable than data outside the cloud. Both of these, again, apply to cloud or warehousing. And more importantly, give us an opportunity for something called data branding, which I think both of our sponsors here have sort of alluded to in their talks before we did this. The idea is to say that there's some data that is of known quality and some data that is of unknown quality and we'd like to use all of it, particularly and have people use the known quality data as opposed to the unknown quality. Success in this area is a truly three-legged stool. Again, if the airlines decided to put another two-legged stool, I wouldn't have a very good experience with that particular flight. Equal amounts and, in fact, larger amounts of people in process than technology, but all three have to be present in most organizations simply rely on technology. Another wonderful set of surveys that I've pulled from over the years. This is from Randy Bean and Tom Davenport. And again, the link for it was right down there. The bottom, but are you incorporating, excuse me, are you driving innovation with data? This is the first year in eight years of survey that we've actually gotten a greater than half. The rest of these aren't so good. Are you competing on data and analytics for in 10R? Are you managing data as a corporate asset, as a business asset? Again, four in 10. Are you creating a data-driven culture? One in four. And are you forging a data-driven, excuse me, data-driven organization is one in four. Are you forging a data-driven culture? One in five are doing this. But the most important part of this survey is that they've asked the question over the years, are your problems largely technology-based, or are they people in process-based? And you can see here in 2018, it was 80-20. Again, these numbers are varying slightly, but you can see the overall direction is still largely around this. All the way even to 2023, 80% of data challenges are people and process-based data challenge. And data governance is the only resource that you have in your organization devoted to addressing these kinds of challenges that are in there. Yet we have this other problem where we try to communicate with things. Now, we're going to talk a little bit in here today about data security and governance and personalized environments, things like that. But the rest of the world just doesn't really get it. You may have data management and try to explain to people the difference between data management and data governance. And you know what they hear? Again, thank you, Charlie Brown from that wonderful little soundbite that's there. Don't talk to them about these things. Talk to them about your data program. Make sure that they understand that a data program is something that you need to have in your organizations and that really all of these pieces fit into it. They'll be satisfied with that. They understand, trust me, they do understand this. But if you try to give them too much detail, they are not going to get it. So while we're not making lots of progress, let's talk about what the actual pieces are. And this is a slide that I love because of course it has Alice and the Cheshire Cat on here. And this is what happens with many organizations. They come into this and they say we want to move our data management program to the next level. But of course, if you don't know what level you are on at the moment, you don't really have any idea what the next level is going to be. If you're currently managing your data, but you can't measure it, how can you manage it effectively? How do you know where to put in time, money and energy into these things in order to support the mission better? Now, I had a title when I was working for the Defense Department that was fancy. It was U.S. DOD, Reverse Engineering Program Manager. And during that time, we sponsored some research at Carnegie Mellon University. Sorry, Carnegie Mellon. I abbreviated you guys at their Software Engineering Institute. Again, a very fine organization that I've been associated with for many years. Asking the question, how can we measure the performance of DOD and our partners in here? In other words, can we look at a process and say, are we getting better or worse at it in order to do this? I also was told, go check out and see what the Navy is up to and found out that's where I ran into John Zachman the first time and Clive Finkelstein, both of whom became my mentors in this process. The SEI, in response to the DOD request, responded with an integrated process and data approach. But DOD required the SEI to remove the data portion of the approach because they said your name is a software engineering institute and so you shouldn't be talking about data. That portion that got cut off was literally lying around at SEI and somebody offered it to me and this fellow here, Bert Parker and I, put together a series of internal research projects for the MITRE Corporation based on this particular model and a couple of grad students that were involved in all of this, coming around and understanding the scope of data management organized into five key practice areas. While this was good for an initial approach and again, the references on here on the slide, you're welcome to go back and read it or just ask us. We'll add it onto the slide deck that gets sent out. This wonderful individual, Melanie Mecca, who's been a dear friend for many years, really took the initial work that we had done and professionalized it and so she spent a lot of time looking at these things and was the primary author of this thing called the Data Management Maturity Model. Now, the first release was done almost 10 years ago. It had wonderful sets of contributions. It has not had additional sponsorship since then, but there may be some things coming up in the future on that real soon because I know there's a lot of people that are interested in it. This reference model framework really specified specifically the maturity practices that are necessary if you want to do this and it did this by putting in a structure for each of these five areas that I've mentioned. It's the core category, it's got a process area, there's a purpose, introductory notes, goals, core questions. It's not exactly scintillating reading, but it is hugely useful stuff and we all owe Melanie a vote of thanks for devoting and professionalizing all of this. Just like anything else in terms of process areas, the model emphasizes behavior. It wants proactive positive behavioral changes so that you can carry out and repeat effective practices and leveraging those practices and extending them across the organization. These activities result in work products that produce processes, standards, guidelines, templates, policies, so that you can reuse these and make your staff happier because they'll spend less time being frustrated on this. The practical effects then essentially show you how to take this and evolve it to an all hands on deck approach and the reason for that is critically important. All of your knowledge workers are doing something with data and most of them have zero training in it in order to do that. Just to give you a little bit of grounding in all of this, this is a wonderful study done by the corporate executive board that looked at various ways of improving processes, I tell Rob, COVID and PMI. You can see for those three on the right hand side actually introducing those frameworks into your organization means that your projects do not tend to finish on budget but with the CMMI there is a market improvement in that ability to deliver on budget and again on time here as well. So these are the only scientifically improvable methods of showing how this stuff works. In fact, I wrote in my data strategy book a little piece and I'm just going to spend a minute and read to you. While all improvement efforts begin with the obligatory assessment phase, Carnegie Mellon's CMMI and DMM are the only proven frameworks that have the added benefit of literally decades of practice and benchmarking data. Organizations not using the DMM risk and inability to meaningfully compare results against other organizations and as a result adopt unproven methods. I can't tell you how many organizations I've gone into and they say, oh, the contractor told me we were using DMM and I look at it and say, no, that's not in fact what they were doing or worse still. I had several of them tell me that they were using the DMM method. Well, the DMM method doesn't exist in this. So once again, it's something that you need to be carefully paying attention to. Let's see how these things work together and I'm showing you this not from where I'm speaking to you right now because I don't have a good enough internet connection here but this is my home in Montpelier, Virginia and you're looking at a picture of my barn. Now you might say to yourself, why on earth are you wasting my time, Peter, showing me a picture of your barn. Well, I do want to tell you a quick story about the fact that I'm what's called a horse husband and that means that part of my dowry and getting engaged to my bride of 20 years at this point was that I had to build a barn for her. We didn't have the money at the time so I borrowed money from the bank and the bank gave me exactly this much money enough money to build a solid foundation for my barn because if I didn't have a solid foundation for my barn, the banks in Hanover County, Virginia where I live understand truly that what would happen here is that the barn built a good barn built on a poor foundation might result in me not paying the loan back in this case because as a horse husband my first duty is to our wonderful four legged friends that inhabit the barn now in order to do this and once again you're saying well wait a minute okay maybe I see this there's a process that you give out a little bit of this money so that people can get started on this but that you've got to have a good foundation. Yes I emphasize those words by striking my fist against the table and the bank did exactly the same thing until I got a foundation inspection done saying that this foundation had been in fact put together the way it should be and was capable of supporting the load that we intended to put on it. They weren't going to give me any more money now this just makes good business sense and I certainly appreciated it from the barn but in particular there is no IT equivalent of this and this is something that we have to change in our practices we have to change the way we educate people about it we have to change the way we talk about it and say that data foundations need to be there before we build IT solutions on top of them or we are taking bad data plus something awesome and coming up with bad results. Now I also like to include a little bit on Maslow many of you remember this from high school Maslow's big insight was that if we have food clothing and shelter needs then we are never going to be safe that physiological needs are a prerequisiting necessary but insufficient prerequisite to being safe around this similarly if we are never safe then it's hard for us to become involved in something that is bigger than ourselves love and belonging to a larger group on this and if we do not have love and belonging in the sense of a larger group it's very hard to know ourselves as well self-esteem is the next level up on this and finally even if you have self-esteem the idea that you can get to self-actualization is again necessary but insufficient if you don't have esteem you don't have the love and belonging you don't have the safety you don't have this physiological means met it will absolutely not be able to work at your highest and best capability well this concept of self-actualization has been hijacked by the wonderful TED talks that are out there they call this flow it doesn't matter what you call it you understand that you cannot be in flow or be self-actualizing if these other needs are unmet and data is exactly the same as this we have a wonderful set of technologies that we put in this golden triangle of advanced data management practices notice i've dropped mesh into their crypto ai and ml and all these other things these are just technologies and the problem is they are the tip of the iceberg when it comes to understanding data and that the foundational data practices the five things that i've mentioned a couple times already so far are organizational capabilities those capabilities must be present and practiced at a high level of maturity in order to make anything at the top work at all now even though i showed this picture lots and lots of times on a regular basis i still get asked the question great peter but i want to do it faster and i say well of course you can do it faster but if you do it faster it will take longer if you do it faster it will cost more if you do it faster it will deliver less and if you do it faster it will present greater risk to the organization so it doesn't matter what you are doing in order to do this you are going to have to do it in a way that provides a good foundation and gives organizational capabilities before you make significant investment technology in fact the ratio for this should be four to one if you're going to invest a million in some sort of technology i will tell you that you need to invest in most instances uh four million to make sure that that technology is fully utilized by the organization to its full capability this is the main source of frustration with most of our vendor partners that are out there is that they can't get people to understand how to use their products they think it's some sort of magic and of course it isn't uh in order to do that let's take it the next step and notice i'm flowing across those five pieces this is back to the cmmi research that i was telling you about and let's just define them if we're going to do data management strategy it means we're trying to manage the data coherently we don't want to have a thousand different knowledge workers trying to implement a thousand different components of data strategy we need to do some things before we do other things that's the essence of strategy around there and that those data assets can be governed professionally we've now been working in this area for well over 20 years and have a class of data management governance professionals that we can put in place and employ in order to do this and we need to make sure we understand the data lifecycle and and that manage the data lifecycle properly in order to do this and that we have data that is fit for purpose uh that's our definition of data quality in order to do this with of course the right architecture technology stack whatever it is that you're going to refer to this and supporting practices from the organizational as well now i mentioned uh earlier on that this is a weak link in the chain method and it certainly is i'm going to give you an example of first of all the cmmi rating on this you get one point for having a pulse that's not a very high bar but nevertheless it does give you that one point it gives you two points if you have the ability to have a repeatable process your repeatable process may be as simple as give it to guy who happens to be an excellent data manager guess guy i'm picking on you for this one i'll pick on nick in a minute here right and then if you get three points you have now had guy right down what it was he was supposed to do in order to do it so that if guy wins the lottery and takes off somebody else can come along in his place and repeat those practices i've worked for some organizations that had all that expertise tied up in one individual which again the more that individual has and the more likely they are to play the lottery the more likely you are to have all that wonderful information go out the window the fourth level of this is that we take that defined practice and we measure aspects of it how many how much what's happening at each of its different piece and finally if you use those measurements to improve your existing practices you now are awarded the five points of optimization we'll see some scores on those in just a minute here but let me just rate each of these things as a three so data governance is a three data quality data operations data platform by the way if those are the three numbers that come up there if you do get rated at those levels you are way above typical in this case however in this instance hypothetically that i'm managing it up i've given the data management strategy just a one meaning that there is nobody who's doing anything to help prioritize these areas and therefore the overall effectiveness of this system is a one that is a very disheartening piece of news because it shows that you must invest with a balanced approach in order to do this. So we've got the assessment components now on the left hand side here you can see the little icon that i had on the front screen and i'm looking at these five data management practice areas and on the right hand side of the screen the five levels that i've spoken of when a maturity assessment is done you might end up with a very nice diagram like this that tells you what's actually happening i'm going to discourage you from investing lots of money in this i'll tell you why in just a minute but nevertheless it is still a pretty insightful set of diagrams here showing that some areas had higher scores than others and notice again very few of these scores exceed the three line the circle that's dots in the middle of that that is again not even close to an average score. Here's another thing that we can do with all of these and CMMI and Melody and others have put together several benchmarks where we can look and say hey how are you doing with these various areas and things like that. Now the idea with assessments here is it's generally not worth a lot of money for somebody to tell you that you're at the beginning of your journey it really isn't but you can use an assessment like process to quickly and and expensively uncover previously unknown pockets of excellence in your organization and the first plan should be examining the feasibility of taking those pockets of excellence and expanding them to other parts of the organization. Let me give you a couple of concrete examples on this here is one that we did at one point in time for the insurance industry and you can see here that the average insurance company at that point in time was not really doing very well these are relatively low numbers in this case here's another example for an airline that i did and you might wonder as did their executives why am i listening to somebody here who's talking about ones and twos and then i was able to say well this is how your organization responds in these practice area assessment but here is your competition and you can see they are behind the competition in this case that's the kind of message that corporate boards need to understand and what the impact of that is going to be on their business we can also add in all the respondents and show that the averages here were pretty much average of all the thousands of companies that we have done doing this kind of practice area one last quick example here oh that's i'm sorry first of all clearly the challenges here are to take the ones and make them into twos first because remember if i take two of the two ones and make them into a two i still only have an overall score of one as in my data practices are being held back constrained by this weak link in the chain here so we have to fix all three of these ones to make them into twos in order to get a overall score of two and more importantly the score is not the important thing the question is what's actually happening in our organizations with respect to their data management practices here's one that i can give you the example of you notice it says world bank and this was the international finance corporation part of the world bank that actually hands checks for large amounts of money to different people around the globe for all sorts of good purposes in this case we're going to look at their treasury group you can see their treasury group was not very high they scored a flat one on this as well as their information systems group at the time also scoring one but the international finance corporation was actually doing world class practices at the time so here is a case where the international finance corporation business people were able to show the treasury group and the information systems group how to do this process better which is a much cheaper process in general than bringing in external consultants you can see they were quite above the average for the banks all the way around and also for the overall benchmarks again very very interesting pieces in order to look at this we haven't done one of these surveys for several years this is one of the other challenges that we have in that nobody wants to invest the money in order to come up with these things but this was still a very valid survey from oh nine excuse me seven to nineteen so a good 12-year period here coincidentally by the way this coincided with the uh the hype around something called big data some of you may remember it some of you do not uh turned out we were never able to identify big data period end of story but that we were in fact able to benefit from big data technologies now big data technologies doesn't sound quite as sexy as saying big data all the time but nevertheless it was a much more realistic portrayal of what was happening here and of course the key for this is that from seven to nineteen these numbers did not change at a time when an amount of data was exploding and continues to explode in fact if we look at literacy overall in our organizations the US government has been surveying these things for several years there's a couple of uh different labels for it if you go back far enough you'll find it as the national assessment of literacy in adults or adult literacy and then they change the name to PIAC and I don't program for international assessment of adult competencies but the point is these three numbers here that they've rated uh citizens of the united states in this case from a literacy a numeracy and a digital problem solving approach also did not change from 12 the last time it was done to 17 now that the pandemic's over we may be able to get them to do it another time just to see but not improving is by itself a problem it's kind of like being in business and not understanding that you need to move ahead always that just simply treading what are in business is not going to be satisfactory under any circumstances let's move now to the idea of strategy around this now strategy guides work group activities it is one of the things that a work group shares a similar shared understanding of what the organizational priorities are and while you'll see lots and lots and lots of books and things out there on strategy I don't like most of them because really the word strategy originated in the military and the business consultants picked up on it and said great I get paid by the PowerPoint so I'm going to make lots and lots of PowerPoints around this I'm being somewhat glib with this but I also see this as many organizations that have you know 100 page data strategies that simply sit on a shelf and are not useful in this if we look at the military context around this it's a pattern in a stream of decisions so what are we trying to actually achieve here so that at the knowledge worker level which is where your data is managed for the most part in your organizations they understand the right thing to do whatever that happens to be for the organization again our theory here is a theory of constraints again ellie who golderot put this piece out but it's idea of finding something in your organization that is constraining you from getting from an organizational level one to two or two to three or three to four if you're really fortunate in that and that you need to make a quick employment exploit that constraint so that you can try to bring it up most importantly though you typically realize that you have to subordinate all non-constraints so that the constraint is fully exercised and that you put as much through it as you can by the way this is all based on a book called the goal if you haven't read that book by the ellie who golderot on this it's a wonderful analogy for everything that we do in data management and of course once we've alleviated that constraint we don't stop we move on to the next one and this just hits a point here that nick was saying you've got to have a continuous approach in order to do this you're never at the top of the process and that you do want to continue to do these all the time let's take an example where the good guys are on the left and the bad guys are on the right hand side i might use one type of a strategy if i was playing capture the flag or war uh in order to do this however i might use a different strategy if we happen to be the kings of the hill and the bad guys are down here in the valley or vice versa if the bad guys are on top of the hill i'm going to use a different strategy than the good guys so the reason we understand this is because one of our previous presidents white eisenhower said in preparing for battle i've always found that plans are useless but that the planning itself is indispensable the idea that you're going to go consult a hundred page or a hundred slide deck to find out what to do next in your data strategy is ludicrous you're not going to know what the enemy is going to do by the way the enemy is data debt here and ignorance of data practices data illiteracy uh in this case so let's move on to a final component here in order to look at this which is something i call a data sandwich and that is the idea that all of our organizations are trying to leverage high performance automation and they're doing this with uneven amounts of data literacy uneven amounts of data supply and uneven use of data standardization let me give you just a quick example of a very very common use of data strategies here in the united states one of our previous presidents george h. w. bush the first bush signed a law that said we're going to use standard labeling for nutritional information of foods and our american population has gotten quite good at looking at a label on a can and saying this has the amount of fat that i don't want to consume and i'm going to change that particular approach a wonderful example of putting in data standards in the organization and 15 years later we've had a wonderful set of successes where people understand this so how does that apply to you individually well you're going to have to start to smooth your data literacy get people more literate about this and again the idea is not just your data people but all of your knowledge workers need this that you're going to have to start smoothing out your data supply perhaps using some of the tools and techniques that nick and guy described earlier here and we'll come back to it in just a little bit and and employ limited use of data standardization because if i can't get these three things to work together i will absolutely not be able to do this notice a wonderful debbing quote on this i don't think i gave him credit but it says this cannot happen without investments in engineering and architecture in fact i went all the way to india at one point to see this sign hanging on the back of this tea farm in india quality engineering and architecture work products do not cannot happen accidentally now the fact that they put that over a cash register at a tea farm in india tells me that they understood that and of course if we add the word data in there it obviously still applies in the major sense of all of this so let's take a look at the other component that we need to have in this and this is something that we in dama international are very proud of we used to be that if you were trying to figure out what to do you had to look at a book like this which was the handbook on data management information systems instead what dama international did through wonderful efforts on our volunteers is create this icon for the dama dm bach we are now at version two and version two here you can see describes a number of practice areas that we have in order to look at this these practice areas of course are critically important we did not do a good job in one sense that what we really should have said here is that these are things that can make up we said these are things that do make up so people will come along here and say oh well the dama dm bach says i must do dama excuse me document and content management uh no that's not in fact well we wanted to say we wanted to the data and content management are part of your overall data management practice and could be more or less important depending on exactly what's happening within those areas similarly we also didn't show any dependencies in these areas the idea that it's probably a good idea although we did put it at least at the center to put the data governance in place here and i'll show you how that works in just a second i'll also note just for the record that twice in a row we've made the data quality chapter the last one we need to correct that in the next version going forward just so that people don't understand that that data quality is last in order to do this so when you're looking at these areas what you see again is that three-legged stool that i mentioned before most of the time you're not going to be working in just one pie wedge but you can't do them all at once so it turns out the rule of three becomes very very useful you may in fact start out by saying i want to build some data that goes into a data warehouse so that i can collect it from our legacy transaction systems by the way definition of legacy is any system that's in production right now there's no point in arguing about it let's just call it exactly what it is the idea of doing a data warehouse without simultaneously trying to approach data governance and data quality at the same time is simply ludicrous notice also that the organization gets a one x participation point or experience point in that area in order to do this and going on a little bit further the same organization might discover that while we're doing data warehousing and governance well so we're getting two experience points in this there was a deficit in the organization around metadata management and so that we need to have to shift our efforts from strictly focusing on quality to instead focusing it on metadata management in that area a third version of the same thing might now say okay we started off with data quality and metadata but it turns out what we really need to do is look specifically at reference and master data management so the organization here has gotten three experience points for governance and data warehousing we would expect those practices to be more mature than reference and master data management metadata management and data quality management which we've only done one experience points in order to look at this we have a lighthouse metaphor that i love to use on this which is the idea that you have things that can further the organizational strategy as you're trying to say what can we do with data but did you also have some data that you understand can be improved for use by the business to achieve business objectives that intersection between those two points on the Venn diagram is a great place to do it but we can also add one more qualifier if you will or one more slice there and that's the opportunity to practice needed data skills and when you have all three of those lining up in one place that's really your golden triangle that you want to start on for your next data management process and again the question i asked earlier on how does one get to Carnegie Hall well of course the answer is practice practice practice but we need two things in order to do that we need some good music you know i'm going to not make the claim that stay in alive by the bg's is a great piece of music i actually hated that song more than any other song on the planet in 1977 when i graduated from high school but when i saw bruce springsteen do it i realized it was a great song just done in a discus style the reason it sounded great from bruce springsteen is because his band practices practices practices as long as i've been doing this that band has been practicing as well coming up with some great music around that all right well bit now here as we move towards the last few minutes first thing is to understand that all of this involves human beings i've said several times people in processes are critically important in order to do this and we have a definition of a profession called change management and leadership that helps the organizations understand this similarly if you're going to become ready for this particular process and i look at the organization and i see it has vision and skills and an incentive and an action plan but i see frustration i know that they don't have the resources similarly if the symptom is anxiety i know they don't have the skills and you can fill out the rest of this chart for yourself only when all of these things line up just as the five parts of the key that i'm lining up here in order to do that do you actually get change so it's not just data it's not just technologies but it's a culture in the organization itself and culture is the biggest impediment to shifting organizational thinking about data i have a full case study on this that is free for you to download you can click on the link when you get the slides from Shannon around this but we're really talking about here are some very big changes the volume of data is increasing much faster than we're able to process it the data interchange and overhead are all costs that measurably sap organizational productivity and that the reliance on technology alone approaches has not material addressed this gap over the last 20 years finally there is an industry type whose sole purpose is to extract data from citizens and then use it to make money so being defensive about the process is absolutely critical process is more important than results you're going to have to start practicing i know if you've ever been in a house that a musician starts to learn how to play something it sounds terrible at first but it will get better if they keep at it that failure in itself is a lesson in order to understand how this works that we need to learn from these mistakes it's not just that we make mistakes but we learn from those mistakes and that people in process aspects are not receiving enough attention as we try to improve our data but that best practices in fact do exist and again if you are working for the united states federal government and not using best data practices the penalty for that is actually higher than a HIPAA violation that usually gets people's attention right away in there so practicing data management better means we have to understand that we are unsatisfied with our current state and almost everybody will agree with that but that we haven't been making progress by simply throwing technology at the problem that we need to understand there is a good basis for this and that this basis it does have a sound scientific proven results that we can do this better than we would do with other approaches and this is the push in the industry for best practices around that there's a comment not too long ago that said best practices mean you can only be mediocre but remember most people aren't most organizations aren't even mediocre in here so I've talked a little bit about the DMM and the DMBock so that you understand how to apply them together to address the weak links in your chain with a little bit of strategy there's more on strategy that you can do and that it is a combination of people process and technology that requires lots and lots of practice in order to do this just to let you know Shannon and I are starting our I think 14th year of these webinars in here it's been a wonderful partnership Shannon over the years and with that I will turn it back over to you Peter thank you so much for another great presentation it has been I can't believe we're it's almost 14 years now we're going to be working on here and agree that has been a great partnership if you have questions for Peter or for Nick or for guys feel free to put them in the Q&A panel of your screen and then answer the most commonly asked questions just a reminder I will send a follow-up email to all registrants by end of day Thursday with links to the slides and links to the recording so diving in here we had a question come in Nick on your question just precisely self-learn yeah so that's a that's a great question precisely is data integrity suite is built on a shared data catalog across all of our services so it does incorporate specific machine learning ML capabilities that that learn pattern patterns and habits to help teams deliver like more accurate consistent contextual data to the business so the real value of that is yes we can garner data intelligence that includes recommendations and alerts that are machine learning based awesome thank you and guide for you where does Satori run Satori can run as a hosted service or it can be deployed within your environment so it's it's pretty flexible depending on your current data architecture I love it and thanks for those questions great presentations for the both of you okay and then moving on here so it's it's on the slide before the lighthouse Peter if an organization question wants to do everything from scratch on a new setup what would be the order if cannot start on all segments at the same time well I think both Nick and Guy indicated that most of these it's not appropriate to pre-specify that certainly at least in the dim-bock we talk about data governance being a central component to this but each solution is going to be different from your organization let's let's just take a step back and I'm you know again I'm back from a trip to China in here and yes still all over the world they have these wonderful signs and all the airports that say all the best companies run software x whatever software x is in order to do that if you think about it if they're all running the same software which means they're all following the same business practices in those areas what is left to differentiate these organizations and the answer is the data in their organization so if you might have two companies running the same software and one of them has good data and the other has questionable quality data you will definitely not achieve the results that you're looking to achieve so there's really no way of staying here what you should do as a starting place but we the three of us are looking at you all the audience and saying you are the experts in this area you are the ones that know where the organizational pain is that know that if you could change this aspect of the sales process you would be able to increase your sales by three percent well gosh that's a great number for most organizations in order to do this or you'd be able to lower costs by a certain percentage or you'd be able to address certain types of customer needs that they're not able to meet now so I don't want to say there is no starting place on this but you as the audience know the best starting place and of course if you have any questions more specifically than that give Guy or Nick or I a quick call and we'll be glad to walk you through that next step in the process I hope that makes sense and I hope I don't sound like I'm weaseling out on this but it really is a bespoke solution for most organizations a custom solution. Guy you guys want to add anything on that? No I'll actually you know you kind of hit on the nerve for me there Peter I there's really a couple things it's it's where the data is it's how you're using it and it's how you're pushing it to make business decisions and depending on where your pain points exist whether it's in data storage data access or data decisions that'll really help drive which of these is your weak point that'll also help drive what tools you bring in that'll help drive how you pitch this to upper management but like Peter was saying we we can't tell you where your pain points are. Guy I'm sure you've got a similar approach to security right you don't just go down lock everything down you mentioned you had the ability to do practically individuals individualized approaches to it I'm sure it's within a context of role-based but nevertheless yeah yeah I mean engineering needs different access to data than leadership then IT then support then the data engineers themselves you gotta you gotta find the problem for the solution or vice versa I'm seeing a lot of things on here that if you improve one improves the rest as well and if you improve your data quality you're going to improve your data modeling if you improve your data security you're going to improve your data management it's there's no single tier here that's going to fix everything. But an absolutely great question thank you so much for for asking. Indeed and Nick and Guy feel free to jump in at any time on any of these this is great love this and so with AI being such a hot topic have you ever had any trouble stressing the importance of improving your foundational data practices in order to set your organizations up for success with their future technology projects? You guys want to go first on that one I've got a very glib answer to it but I think it's it's it's appropriate. Well let's first remove a little bit of the marketing around AI right large language models and and gen AI has an attribute right now in the market that's overhyped I think you know data science has been around for a while and using self-learning models to improve that is probably what people really mean when they say AI is coming around is how can we make our business decisions self-driven ask a question get an answer instead of ask a question define a solution build the product get an answer and so there's there's a couple key things I think that are not really looked at and the first one is how is that AI or data science platform accessing your data is it accessing it in a secure way is it storing it in a secure location is it or do you have audits around how this AI is is getting that data or who granted that access is it give me a copy of peter's access or you know nick asking for some stuff because he needs a smarter business decision if you can decouple those questions you can get to the root of the answer yeah I would probably just echo what what peter had mentioned earlier about about you know this is uh just throwing technology at the problem isn't necessarily going to fix it but actually taking a step back and diagnosing okay you know we see like we've seen over the last year or so how many AI tools and companies and startups have come out of nowhere throwing AI every single problem out there and similar to like the dot-com bubble you know only a small small percentage of them are actually going to provide enough value to stay around and so I think that's kind of a approach that we've kind of been taking it is is focusing much more on the foundational data practices and it's trusting less about you know AI ML and being more very prescriptive about how we implement that and truly I am sitting at Virginia Commonwealth University in Richmond Virginia in the School of Business however if I go out of my office and walk to the edge of my Bluetooth range I will be in the computer science department and what I get from the computer scientists that are down there and they're my friends and colleagues and have been for many years is that they'll come around to me and they'll say hey peter have you got any data that looks like this and of course I know what they've done they've come up with a wonderful new algorithm and that algorithm fortunately or not has no utility at the moment they have to find a use for it and of course that's just backwards what we should be doing is as both of our guests have mentioned today is that we should be finding out what are the business problems that we're attempting to solve and then designing solutions for it but unfortunately we reward these individuals with just simply nimnus and nimnus is not good enough as far as that goes and this has led us to several AI winters that we've had if you go back and look over the here it's another talk that I do but if you look over the the period of AI we've had several places where it gets to this peak height as was mentioned and then it falls down into the trial of disillusionment now it's not to say that we haven't done anything good but we've gotten as far as we've been able to with this and where we are right at the moment with respect to all of this is that me my colleagues around the corner and many other fine computer science departments around the world have come up with what they call learning algorithms and all they need is a great set of data to train their algorithm on this new new technology and everything will be great well guess what that data comes from the three of us who are telling you all what you need to put in place in order to do this so that you don't in fact end up learning that the difference between a dog and a wolf is that wolves only exist in snow and dogs do not of course if you're good to see my dog this morning it was in snow so we know that is just simply false so I'll just close out this AI discussion with a couple of points and first one is that your job is not going to your not AI is not threatening your job what is threatening your job is somebody who knows how to use AI better than you do and that's a really really hard sort of a process to learn which means all of our students all of our employees all of our knowledge workers need to be playing with AI and understanding what does it mean to work with a technology that hallucinates at a rate of about 15 percent I'm pretty sure my students wouldn't put up with me if I gave them 85 percent good stuff and 15 percent stuff that was absolutely complete rubbish that I had made up in the process and again it's not that all of it is bad but that's certainly the error rate that we have in the ML community right at the moment one last point on this and I know that Informatica is not a sponsor here but I do have to give them credit their new tagline for this next year is everything's ready for AI except your data which I thought was kind of a nice nice little tagline on that so maybe we'll get them to come on and sponsor a line first one of these days Shannon indeed thank you so so many questions about AI and what's going on right now I love it so but diving in here what industry search certificates and credentials are available based on the DM buck Peter I know you have answers for this one absolutely as part of the efforts again the volunteer efforts at Dama International we have created a certification program that is called a certificate of data management professionalism the basic place if you're interested in learning more is something called cdmp.info out there on the web it gives you some scores statistics you can also go to the Dama website dama.org and take a look at it as well we've had like I said about 4,000 individuals pass through this program and become certified in these areas there's much more detail out there on that but it's certainly a good one and we know it's good because we are now starting to see requisitions even in China last month one of my sponsors came up and showed me and said hey look we are looking for people over here who have a cdmp so it is a certification that is becoming respected it's becoming utilized in more and more places and more and more job candidates are being asked to have this certification in order to do it that is not to say that you will know this entire wheel by the time you're finished we're giving you a baseline of it and that you can go on and get additional levels I wouldn't make too much of a commercial hero on that channel but thank you for the opportunity we always like to have more certification done as opposed to less indeed and I said dmbock and just to reiterate for those who are not familiar it's a data management body of knowledge it's our common our common acronym that we that we refer to it as um and when will the next version of the dmbock be polished Peter do you know the question that everybody wants to know the answer to well the the answer is it'll be done when it's done but we are in fact organizing to do this we've hired at dama international as one of our first contractors that we've been able to bring on somebody who does have experience with building these out so while the first two versions were done by volunteers we're not going to have a professional approach to the next version of it and there will be some announcements in the near future very nice and it was not that long ago that 2.0 came about so 2017 is there any generalized statistics statistics on where companies most commonly struggle within the dama framework let me again toss this one back over to nick and guy and say from your individual perspectives where do you see the most problem areas now go ahead and put the framework back up there so that everybody can uh see what we're referring to who wants to go first nick um i'm actually going to pass it over to guy i'm i think about it for a second that didn't mean to put you on the spot no worries yeah so you know from our perspective at least at satori we find that data operations and architecture are typically pretty set you know you get you get decisions made at this is where we're going to store our data this is how we're going to store it um operations of the etl space and um and downstream where we're finding at least the most the hardest part to get change is in governance make a new regulation came down and we have to delete users from access and you know deleting a user doesn't necessarily mean deleting them from the database that it could mean hide them from responses it could mean a number of different things and how you manage these the assets how you how you govern your data um tends to be a really hard problem to solve if you don't have the right tools in place or or at least the right mentality to get these tools in place not not saying use satori necessarily obviously that'd be pretty cool from our perspective but there are there are a ton of solutions out there whether they focus more on compliance or more on enforcements of this compliance where how you manage your data assets can really start empowering speed to data product and an example I like to use is in the healthcare industry they've got an immense amount of data that is super protected by regulation at least here in the united states and that doesn't mean that a organization who's hosting healthcare data can't use it to gardener business decisions they have to just be very very careful with how they access and expose that data to the policies and so that from at least from where I sit and the customers that I touch a lot implementation and definition of governance seems to be a huge problem I forgot to mention so nick and guy I saw that you put in the chat because there was a request for your contact information so if you all didn't see that you can scroll up and see the their links there and of course I'll send that out in the follow-up email as well so moving on to the next question here so over the past 20 plus years I've discovered and shared many documents with leadership from the sources mentioned here and every instance they have dismissed as a nice to have or we don't have time for that have I somehow misfinding the silver bullet all these years of you know sharing this with with management the real key with this is a short feedback loop and demonstration of benefits in some organizations that demonstration of benefits that is going to look very different from other organizations so I once went into one organization and justified an entire investment in technology and people in practice issues based on the fact that they were turning over their sales assistants every 90 days I'm sorry so sales assistant sales analysts every 90 days and that that company cost that the company cost to replace a sales analyst was $50,000 so imagine this terrible scenario you've got somebody that's supposed to be telling you where to put your products in the marketplace and they only last 90 days on average and that even when you do that it still costs you $50,000 to replace them well that set of costs right there was more than enough to justify the entire initiative around something that should have been obvious which was that they had four unintegrated sales systems that were telling them where to put their products and that the sales associates sales analysts were extremely frustrated because they spent four days a week integrating data and not analyzing it and of course that was not really the job that they felt they had been hired in order to do but Shannon before we lose the the sort of silver bullet approach I agree with the idea that governance tends to be a weak link in the area probably because it's the most recent of these that we've done and on one of your publications uh Data Diversity publishes a wonderful piece called um oh goodness uh P. Dan the data administration newsletter and two of your other stars in the field John Ladley and Tom Redmond published an article recently saying that data governance had failed I don't know that I necessarily agree with them 100% I've seen lots of instances where it has succeeded but I think they're speaking generally which I don't know that they are inaccurate on that on that and I will try to include a link to that article in the pieces when we send out to you because it was kind of thought-provoking and both of them like to be a little provocative so no no problems from that perspective on that. Nicker Guy any you want to add anything and have any you found some success in working with your customers and talking to executives and what helped get them on board? I'll just double click on what Peter said it's you have to find the pain point for that leader you know a a chief revenue officer is not going to care about an improvement in technology they're going to care about cost spend and how can they improve their bottom line a CTO might care about time spent by a team operating on on tickets a VP of data engineering might care about the amount of time it takes the data query to return so you can't you can't come with the template and say oh I have a minor framework that we need to follow because it's industry best practice you have to find the thing that that leader particularly cares about and and make sure that the solution you propose has at least some mapping to that yeah just I guess marketing yeah you call it a triple click um it really is I don't think a silver bullet here I think you know that's kind of why precisely is is really uh preaches this this kind of business first approach um which is just really abstracting um like no one buys data governance for the sake of data governance um but really because it solves like a business problem behind it right so backing up and trying to spend some time trying to figure out what is the business problem that that is data governance for example could address um and and that's one of the reasons why our strategic services team is is really so valuable uh since they've been there and done that for you know over 50 plus years they use this proven methodology to to bring experience that kind of source as part of that of that equation out that's you know so difficult to to really figure out in the first place great question I love it yeah um so we can probably do a whole webinar on this next question and that and we have a little bit so but doing data better can you explain the role of data architecture and data governance and good data management the main thing that most organizations forget when they're talking about architecture is that architecture represents things what those things do and how those things interact with the other things that are part of the architecture now that's as about as abstract as you can get and maybe people unfortunately stop right there but there's one additional component and architecture has a purpose so the purpose has got to map up to faster better cheaper or less risky and by the way not all at once because that's nirvana and none of us actually get there in this particular lifetime so there has to be an emphasis you have to say we are going to emphasize in this architecture speed uh or or again security or whatever but but you cannot do all of it and if you don't do any of it you have sort of a bland thing that's not really helping your organization in the best way that it could again great question again guy nick what you want to add on that uh go ahead nick um peter can probably probably does the best has the most comprehensive answer here but you know i would say that just doing data better will always be a combination of people processes and technology and so um they must exist together in order to kind of really drive home that what precisely calls that data integrity there's the ratio of again i get this a lot i'm not sure i hit it very well in the presentation but let's just say somebody comes along with a million dollar investment my guidance to them is if it's a million dollar investment in technology you're going to need four million dollars in order to make sure that that technology is utilized fully and best and so it's best to think about it if somebody says i want to spend a million on it to say where's the other four million going to come to make sure this technology is successful for our organization and if the answer you get back is no i've only got a million dollars then i say well you need to scale back your technology investment to about two hundred thousand dollars and put eight hundred thousand dollars into the reinforcement aspects of it very nice so what would you say is a basic understanding our knowledge worker needs to know with their eyes with their eyes glazing over the key of course is not to glaze their eyes ever most knowledge workers do not need many courses and statistics and other things in order to understand all of this but it really does come down to basic data literacy components and it is so important in china i just can't resist this i know i'm showing a screen here but let me just pop up a quick little piece that i was looking through earlier here to show you how much most organizations are actually paying attention to this and the where is that there we go so here we go this is how much the chinese people think it's important they made an enormous book i call this the big print edition of the book but nevertheless it is still quite clear that if we don't take society in general and improve the literacy around this we're really not helping because then it just beats it's seen as a technology problem and it's got to be viewed as an overall societal problem in order to come up with this this is probably getting beyond guy and nix remit i'm sure you guys are trying to change the world as well but this is this is really how it's going to do if we can start to put these types of concepts in place to say that everybody needs to get more data literate and that is as important as having the right technology in order to do it again having electric car is great but if you don't understand recharging on electric car you're going to have some problems with it yeah any thoughts on that yeah i think just kind of focusing on um helping them the you know the amount of time to understand how you know for example i keep going back to data governance but how for you know can how data governance can benefit them um they don't necessarily need to know the entire scope how it works from right you don't need to dive into how tesla batteries work in order to charge it up you just need to know enough to uh tell them how to plug it in and unplug it yeah and remember what other part of your organization is going to even be addressing these kinds of problems assuming the organization recognizes that those are the types of problems data governance is the only resource there is no other component in your your hr departments not out there teaching people how to be data literate uh at the moment it's just not part of their remit indeed so many uh questions around communication right just learning who is your audience and how to best address them um we've got about four minutes left so i'm going to try and get in as many questions as possible so can you reiterate why you would choose to address only three practice areas when setting up your data strategy why not try and define all knowledge areas in the strategy well of course how much do you know at the beginning of your journey it's like trying to say how much is a how long is a project going to take and going to cost right it's just a very difficult proposition because these are new areas and most people who are involved in them have not done it once nick mentioned before the that uh i think it was nick that said his organization of 50 years of experience that they brought to this that is a wealth of experience that most people haven't had a chance to grow and learn from again one thing about being old is that you've seen a lot of stuff and hopefully i've learned some things from that process i'm sure my two colleagues will agree with that as well the idea of three is that it's generally a good rule to say that you know we should not try six at once uh in here and that if we do i've seen groups do four you know but three seems to be a great number to start off with and then that orange circle in the center seems to be the part that organizations are struggling with the most and notice of course it is kind of central to the overall model central to the overall piece so yeah if you try to do this there are 11 pie wedges here or one circle and 10 pie wedges if you want to get technical about it but how are you going to become good enough to figure out what you're supposed to do in each of those areas right from the start it's a very very difficult process so get good at a couple things not all organizations are going to be dependent on document and content management not all of them are going to be really really dependent on purposely developed data storage and operations management yes everybody's going to have them but if you're going to really need to have supercharged access the way microsoft likes to advertise their cloud you know available everywhere ubiquitous instantly um hey you know that's going to be an important component so once again i'm turning the question back around to this wonderful group of hundreds of you that are out there and saying hey uh you know you are the best experts in your organization as far as knowing where you should start and what sorts of things you should take off onto again i'll let the guy nick jump in here as well please yeah i just want to say that it's also important to not really think about these as singular vectors you know where peter recommends starting with three and you know we all of us would argue that more is better and and than less but solving one sometimes inherently solves others and when you do solve these problems you want to also think about them dynamically now don't don't set things in stone that are going to pigeonhole you into other solutions or problems later if you're creating a data model that can be used for an ai prediction algorithm you know make sure that the algorithm is unlocked into this model that it can adapt to other things or if you're writing a data security management standard you know can this standard cohabitate with other architectures or other data stores solving one solves multiple almost always excellent point yeah i would i would just agree with with peter we have a lot of instances with precisely where a lot of companies will talk to try to catalog all of their data only to really find out you know from our understanding that um only really five percent of it is critical to uh to decision-making and so um you know starting small improving value is really important um and that's kind of the whole you know and when i talked about in my presentation the the whole piece about you know starting exactly with with where you experience the most of my pain right um i know peter had that slide out that talked about like if you out of those five categories if two of them are two out of five but three of them are one out of five well overall you're still one out of five so you know that's the weakest link so working on those initially uh and then starting to build up everything as you continue to grow perfect well thank you all for this great presentations and discussion it has been such a great webinar thanks to all of our attendees who have been so engaged in everything we do we just really appreciate all of you and again just a reminder i will send a follow-up email by end of day thursday for this webinar with links to the slides and links to the recording and everybody's contact information so thank you all so much but and i really appreciate it thank you thank you real pleasure working with both of you i appreciate it thank you have a great day everybody thanks all