 Hello, and welcome. My name is Shannon Kemp, and I'm the Chief Digital Manager of Data Diversity. We'd like to thank you for attending today's DM Radio Webinar, the importance of data governance sponsored today by Calibra. It is a deep dive into continuing conversation from a DM Radio broadcast a few weeks ago, which if you missed it, you can listen to it on demand at dmradio.biz under a podcast for a proud producer of the DM Radio event. Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. If you'd like to chat with us or with each other, we certainly encourage you to do so. Just click the chat icon in the upper right-hand corner for that feature. For questions, we will be collecting them by the Q&A section 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 dmradio. As always, we will send a follow-up email within two business days containing links to the slides, the recording of the session, and additional information requested throughout the webinar. Now I want to introduce our host and analyst for the day, Eric Kavanaugh. And if we're ready, there we go. So let me turn the webinar over to Eric to introduce today's webinar and today's speakers. Eric, hello and welcome. Hello and welcome once again. Thank you so much for that great introduction, folks. Welcome to the DM Radio Deep Dive, the value of data governance with Calibra. Very excited to talk about this issue today. And I can tell you, we've discussed data governance for years. Obviously, many of you know that. Many of you regulars out there and practitioners know it's been a topic for many years now. But in just the recent couple of years, it feels like we've had significant movement and there are real changes, very positive changes happening. And part of that, of course, is coming from regulations. So let's dive right in. You heard about Daniel and she has some laryngitis today. So we'll be hearing from Basiliki Nicolopoulou. She's a field engineer, so she works extensively with their clients on this. It's always nice to get someone who has that frontline trenches experience working with clients. So we're going to be hearing all about her experience and expertise in this space and how it can benefit you all out there. So here's my big somewhat rhetorical question. Can data be governed? Well, the short answer I would say is no, only because governed things have to be animated in some way, shape or form. You can govern people, you can govern processes, engines, decisions, machines, et cetera, but data itself arguably cannot be governed. So in that sense, I think the concept of data governance is perhaps a bit misleading, which we really need to focus on is governing these processes and setting up rules and policies and procedures using technology, of course, to facilitate that such that you can reasonably implement them and get some kind of governance on what people do. Because of course that's the key, right, is what do people do with that data. We also have topics like IT governance that we've talked about in the past. Today we'll focus more on the governance as it applies to data, but we're going to try to focus on how these different practices can help your organization beyond just remaining compliant with the various regulatory agencies. Of course it's good to be compliant, but as we'll discuss here in a minute, it is challenging, it is difficult, but the bottom line is you want to make sure you get value from these things. And I recall back when Sarbanes Oxley came out, that was one of the more clever practices that spun out of that act, which was organizations, consulting firms in particular, working with large companies and helping them not just become compliant with Sarbanes Oxley, but helping them really better understand their businesses, better understand their practices, and then be able to change things. If you don't know how things are done in your organization, you're going to have a really hard time changing how they're done, so it's really important to understand what those processes are and how they operate, and the bottom line is that's a bit of a challenge, so we're going to learn about that today. So GDPR, you've heard all about this, I'm sure, the Global Data Protection Regulation. It was adopted by the EU, but it does affect companies in the United States any time they leverage data from someone who lives in the EU. So EU citizens' data is what GDPR focuses on, and I think what many organizations are realizing is that this is kind of a big straw in the wind, meaning things are changing here, even in the United States. Now, granted, we have rules, laws like the Can Spam Law in the U.S., which means simply put, you can spam. That was one of the more well-defined or named, I suppose, laws in the United States in recent past, because it is what it says it is. But the fact is things are changing even here, so we're starting to see organizations really appreciate the fact that they have to understand where their data is, who's using it for what purpose, et cetera. So that's all really important stuff, and I think it's going to be driving a lot of innovation over the next few years, and I really have a very positive outlook about the end result of all this. And of course, at the core of GDPR is privacy, is respecting someone's privacy, is respecting someone's data. Let's talk about some of the specifics and how they can affect you. So these are some of the really important provisions in GDPR, the right supportability. That basically means someone has the right to take their data, which you, ABC Corporation, hold, and port it to someone else. So let's say you're changing your phone company or changing your internet provider, or changing really any kind of service provider that has your data. What you have the right to do is give that company which has your data that you want to give that data to this other company, to their competitor, most likely. Well, this is pretty ambitious, I have to say, because obviously customer data is held in all sorts of different formats, in all sorts of different systems, in all sorts of different ways underpinning data models. A lot of stuff is just different. A lot of it is in fields, just in text fields and notes fields. Well, anyone who has moved data at any level of scale knows that's a very challenging problem. It also has this right to erasure, which I would call tremendously ambitious. And just to be blunt, in my opinion, it's unachievable, realistically. It's just not going to happen, especially for a very large organization that has lots of legacy systems, customer data systems, procurement systems, products-oriented systems, ERP systems, for example. Even stuff like clickstream analysis, for example. Imagine how many companies have gathered clickstream data on you as a consumer. Well, there are lots and lots of them out there. And what the GDPR says, and of course it goes into effect in May of this year, is that if a customer comes to you and says, I want my data removed, I want it erased, technically, according to the regulation, the company has to track down all those bits and pieces of data about you and erase them. Well, there are lots of challenges here. One of them, think about this, is that some of the database technologies these days require you, or I should say, some of the database technologies these days do not allow deletion. They just overwrite or append records. So you can roll back, like in case you made a mistake, you can roll back. Well, how does that apply? So there are some real interesting challenges about being able to live up to these regulations. But what I'll say, quite frankly, is that the key is to have a policy in place and to implement that policy in order to adhere to the rules. So I think that, by and large, the regulators are going to be very forgiving on this subject if you have a plan in place and if you are executing on that plan for how to deal with it. But like I say, right to portability, right to erasure, extremely difficult to achieve, but it is setting the bar high, so that's fine. And privacy first, we already mentioned. This one, the next one is a big one. Data breach notification within 72 hours of learning about this, right? Well, think about some of the data breaches here in the USA recently. There were data breaches that went unannounced for weeks, months, even over a year. This, I think, is a very reasonable rule because companies really should be required to announce when something has happened to protect the consumers. That's the whole idea, is to protect consumers, to protect your information. Of course, there have been some massive breaches in the recent past here in the US. There are going to be long-term ramifications of that. I've heard off the record of some really, really intense work that's being done behind the scenes at very, very large government agencies, let's just say, because of some of these massive breaches. And I think that's part of why we see an extended and really robust focus these days on governance and on getting this stuff right. So data protection impact assessments, you need to be able to show that you have done assessments to determine what would happen if there were a data breach. It's like a simulation, basically. You need to understand and be able to articulate to a regulator what your systems look like, how they operate, how you're going to determine if there's a breach, and then how you're going to address that once it happens. So also, some companies, especially companies that have healthcare data or any information about your health or your situation health-wise, they have to get data protection officers. So those are going to be dedicated officials, executives in large organizations who work on that specifically. And as I mentioned, international companies are implicated, and last but not least, 4% of revenue, of turnover is what they call it, but global revenue, 4% of global revenue, I think we all know who these are really targeted at. They're targeted at big companies, but also specific companies like Facebook, Google, Yahoo, all of these really big Silicon Valley companies, I promise you they're going to be in the crosshairs of the GDPR. So this is a very interesting picture right here. This is one of those pictures where if you walk past it, it starts changing all of its different views and colors. And I came up with this idea because I wanted to articulate the value of seeing the big picture, of being able to see beyond the initial view of something and better understand how it works and what it means and what it does and all that kind of fun stuff. So if you go to the theater sometimes, you'll see posters that change as you walk past them. That's a very close-up look of this kind of imagery right here. So I'm saying this because what I want folks to understand is that you have to look beyond just the granular-level issues of compliance. You have to see the big picture. It's really important to understand what this stuff all really means, how it all works and how you can benefit your company beyond just being compliant. So finally, last but not least, before I turn it over to our friends from Kaliber, trust really is the anchor of every business. If you think about how we do business with each other these days, a lot of times you'll do business with someone you've never even met in person. You'll do business online, just finding someone. We really do have trust bait into the core of our business culture here in America. So it's important to understand that trust really is valuable and once you lose trust, it's very difficult to get back. So the key here is in terms of data, understanding your data, understanding your processes, understanding your organization and your culture in particular is really important for being able to build and foster a culture of trust. And once you have that, I promise you're going to do a lot better. If you don't have trust in your organization, you're going to have some real serious problems. And with that, I'm going to hand it off to our friends from Kaliber who are going to share some interesting slides about their take on the world and some good insights about how you can improve trust in your organization and get some better value from governance. Take it away. Good afternoon. My name is Vasiliki Nicolapolo and I'm with Kalibera since the beginning. I worked many years before that with data governance and information management. So the subject today is going to be not so much the importance of data governance but what it is important within data governance. So let me... Okay. So of course I very much like the introduction made by Eric because a lot of the GDPR regulations that came up in Europe lately represent a very nice baseline for how private data should be handled in terms of data governance because they start with policies, they start with processes, it all starts from a business perspective as well. So that sets a very good pace and enables companies to get started with the data governance programs and expand on it. So the presentation today is not about GDPR. However, I thought it was... And of course we can talk about GDPR and we have solutions so we can do special presentations on that but I wanted to make that connection with the presentation that Eric made and continue with the fact that we see from Gardner and from other marketing research that there is an uncapped value, a potential of one trillion based on some analysts, a value that's hidden in data, data that is not properly used. There's a lot about digital strategies in companies, there's a lot about initiatives and programs for digital strategies but a lot of the data, a lot of value that the data, these data contains is not yet exploited by companies and management. So data governance plays that role to help identify what is the value of data and put it to use. So so far the data, the focus of the users for the data right now are scientists, engineers, IT people, business analysts and the people who are neglected in this using the value of the data are the casual users, the business users, the data customers in general. So their needs are not met yet. So what does that mean? That means that business users, general users in general, any user in a company does not really know how to find and understand or there's no way for them to find and understand the data. Find what data they want, for example, a manager wants to create a report about profitability, for example. Where is that stored? How is it being calculated? Where can I find the trusted information? So you not only have to know what you're looking for and where you find it but also how you trust it. And there's a lot about risk management here about the risk of not making the right decisions or wasting a lot of time finding the data. I was working at a company as an enterprise architect some years ago and I remember we had that problem with many business users that were doing self-service BI, for example. Great tools, great technology, but there was no understanding where the data is. How can I get the right data for my reports? So there is the old school and the new school approach to this. It's a new approach to how we manage data. The old fashion is to go via IT. So technically, technical artifacts, it is about technically for technology efficiency, maybe cost reduction. It's very much locked down to technical and regulatory focus. Now this has shifted to more usage focus, making sure the users get the value they need for their business requirements in everyday needs. It's data-citizen enablement empowers the users, the data-citizens, to do things, to change things with the data and incremental value because we don't just write a book about a glossary of business terms in the data dictionary and we'll put it on the shelf. It is continuously improving, continuously changing and growing. So we need to capture the change properly through the proper processes. So it has to be social and collaborative in this way. So the best practice and the success comes from a balance between a good offense and defense. So the old system was to catalog the data, the technical metadata and all that, but we have to do additional, so there was some core governance, but we have to do some additional work on top of that for the data experience for the end users in order to be able to search and find the data they want to trust it as well and to understand the data properly. And so creating a catalog of assets from a business and technical perspective combined. Okay. So how do we identify the measures and outcomes of this? So for trust we have to identify the rules and the business rules and the technical rules and the errors and the issues through issue management, through increasing description and catalogs of rules and business rules and data rules and policies. Define the access, not allowing just every user to access everything. I have seen with customers in the past they were giving just in order to avoid the problems just to give access to users to the core databases. The result was poor performance obviously. So and then we were looking to find how we will maximize performance of the systems where you will never be able to do that because by allowing everybody to access and retrieve massive amount of data just because they don't know what they need to retrieve doesn't solve the problem. So we need to regulate access, with accessing for how long and exactly what and not keep replicating the data on different systems. Quality, it's about again rules and policies and compliance and tracking everything with issue management and with processes. So we need to focus on the value we bring to the end user. And the end user can be the IT user but most primarily is the business user. So we have to focus on the value we give to the business users and who are the primary stakeholders. And in order for the business users to get value out of the data, we need to talk their language. We need to collect information from these business users about what is critical, what are the critical data elements, what are the processes, what are the rules they want to define. So it becomes a distributed way to store the data. And of course it has to do with processes and which are the methodologies, the step-by-step processes to manage change, to manage onboarding new assets, business or technical, and to track everything. And of course we have to be outcome oriented, not just import all the technical metadata we have and just wait and see if that brings any value because it's not going to. I was talking to a customer yesterday and they told me we tried the technical approach onboarding all the technical metadata and it just didn't work. We imported a bunch of technical tables and columns and we didn't know what it means. We don't know what to do with all this. So we really need to measure the outcome of all this work through metrics. So I'm going to start with the approaches. So it has to be very important to be able to be flexible. Here I'm showing you here some examples of implementations of data governance. These are customers, some of our customers who have implemented data governance and they use different approaches. So the approach has to be flexible. So the first case that I'm presenting here, there are many more, but these are the main three approaches. One is the self-service and analytics BI case in which because the user was primarily, the business user initially from the beginning through reporting, the approach was top down. So we started with the reports. We started with business terms and glossaries and metrics and KPIs and we drill down into the technical metadata that are linked to this business information. The second approach was from some technology companies, Dell EMC, but also a financial institution which did the bottom-up approach. So they started with importing, ingesting technical metadata from the data lake they had and the data warehouse and they built it up with context like business meaning and processes and rules around the data that they imported or ingested. And the third is what the regulatory compliance is imposing now to companies which is starting in a parallel fashion, starting from policies and regulations and expanding both on the business and on the technical front, but starting with policies. And I think that's a very healthy way to start data governance. So the next concept I would like to mention is that once we catalog the business and the technical components, it doesn't end there. It is incremental. In other words, every time a user changes something, we are adding a new table or we are changing the definition of a KPI or we are preparing in a different way the calculation or the aggregation of a value or we combine in a different way or a new asset, a new hierarchy and we store new data, et cetera, or we provision. Every time there is a new step going on with the data, it has to be monitored. It has to be documented. So every... And that touch delivers value to the data so because you know what's the latest and what's the approved, what's the approved definition in the current state of your data. So it is not static. It requires change management. So this is an example of a catalog that we have which is based on the IN analytics. So you see here a catalog of all the reports that the business users are using. You can see here that you can have certified reports or assets or data sets. So you can see here this green ribbon here that means that this asset is certified. That's very important, the concept of certification because you allow users to define... to rate the value of the data and not to look for... not to use data that are perhaps not the right data. Again, I was at a customer this summer and they told me that they had a CFO meeting and they were five sales managers in that meeting. It was the same report, but every sales manager had different numbers or different numbers in this same report. So some of these managers, they had to go back at their desks and do the job right because they didn't use trusted data. They didn't use the data from the right source. So certifying the assets, going through the process with workflows that defines exactly what is certified, what is trusted and who owns it, ownership and responsibility, that makes a big difference in reducing costs, reducing risk and increasing value. So the other approach is the regulatory compliance approach in which you start from the policies and the processes and you don't start from reports or you don't start from tables and columns, but you start from what's important to your organization, which is your policies, your processes, your projects and the reason why you're using this data and then you expand this on, okay, what are the reports, what are the critical data elements, what are the data associated with these processes, what is the ownership, who owns what, what are exactly the roles and the authoritative sources and the applications that are using this data. So this is the regulatory compliance approach and it is actually a very good baseline. So in other words, to summarize, we tailor the approach. So the approach has to be flexible for each customer who wants to start from a different perspective. But in either case, in any case, we have to be able to allow incremental management and processes, cataloging for assets like business intelligence and others, and of course, the middle out approach or parallel approach for business and technical with the regulatory compliance. And of course, we have to map the value to the stakeholder and make sure that the stakeholders get the value they need. We have to measure that through dashboards and hit maps in a nice reporting monitoring way. Again, the approach is, of course, you have to have the business cluster, the data dictionary, the lean edge, the data quality, the profiling, the sampling, the search collaboration, sharing, artificial intelligence that helps help desk issue management, reporting regulatory compliance, but also the operating model. So what is the operating model? What is the process flows? What are the policies? Define the stewardship processes methodology for your organization, your reference data, defining the way and implement certifications of your assets. So the stakeholder value is multiple phases. So for data scientists, for example, which is the offensive now, the defensive maybe you can say was the data stewardship, that's the older approach with the definition of the tables and the columns and the systems and all that. But for data scientists, for example, they have to search for the right assets or data sets that create new data sets. They even create new data lean edge. They're looking for profiling. They use machine learning to find and ease that. And also, supported activities are the utilization of new data. They may create new data. They compare data or they profile data and they may create with this data the data scientists, new data models or new analytical models, new visualizations. And of course, the stakeholders could be strategic planning users or the chief operating officer. For data stewards, on the other hand, is workflows, approvals, making sure the tasks are completed, access requests, making sure that there is the proper data access request for end users. So they request for access and they get permission in a proper fashion. Data sharing agreements, not within the enterprise, but also in exchange of data with other vendors, partners or countries that becomes actually regulatory compliance as well. Semantic search, authoritative source identification. So it's not just importing all the data sources. Well, I had customers who were asking me to import, let's say, they had 50 sources for the same business term. Well, in reality, it's good to know what's the authoritative source and of course you can catalog all of the sources, but it's good to qualify as trusted. And metrics again for operational reporting, definition, and usually the stakeholders in that area are the chief data officer and the data architects. So the major metrics to make sure that our stakeholders and data users in all fronts are happy with the value they get is to measure the utilization of the data, measure what data is used, how it is being used, and the ability of these users to create new data themselves and also to get value, new value that was not possible in the past. So the bottom line is that we are creating a new data experience. We are improving the old notions of governance by balancing including new concepts like the operating model, the workflows, the issue management, the stewardship, the catalog, a balance between offensive and offensive approach and a focus, very much focus on real value for the end user, not just for the IT people, although the IT people, I was also a DBA for some time, so even I as a DBA wanted to have a business concept around the databases and the tables I was managing. Even the IT people need that business layer in their day-to-day work, not just the business users. And also the capability to be flexible in the approach by the use cases that top-down or bottom-up or middle-out. So you have to be able to use the approach you are possible with, but at the end, you have to end up with a holistic picture, right? Not just the bottom or the top or the middle, but the entire picture at the end has to be completed. And again, you can start step-by-step, face-by-face. We had very large customers who were telling us, where do we start? We are such a huge company. We have lots of data, but where do we start? You don't have to start big. You can start small, face-by-face. Start with the most important project or policies or critical data elements and then grow. And of course, establish and measure and monitor the metrics and the value to the stakeholders. That's very key, because unless we bring value to these stakeholders, not just to the IT people, but to the business people as well, then we have not completed the task. So with that, I just wanted to also mention we have a great free offering link to which is the university of Colibra, university.colibra.com. It has free courses. Some of them are 15 minutes long. Some of them are 20 minutes long. Some of them are half an hour. You can do them at your own pace. Again, they're free. You have to log in. And they provide great value in terms of giving you use cases, examples. Guiding you through the process and the functionality. And it's not just product oriented. Some of them are product oriented and some of them are more general for the general education. Thank you very much. And if you have questions, we can take them now. Okay. Can we do a lot of... I'll go ahead. All right. Can I just chime in? This is Dan Shuller. And I'm sorry you could hear from my voice while I asked Ms. Ellicke to join me here. I just wanted to say real quick, you know, this notion, we ended up with the concept of metrics. And I think, you know, this idea of the split between offense and defense is something that, as we approach data governance, you really have to have front and center these days. You know, the defensive data governance, the way you knew you were doing it right was, you know, nobody got sued. Nobody went to jail. You know, that kind of thing. You know, we weren't breaking the rules anywhere. The offensive data governance really is a focus on how do I make the data valuable for that broad set of people? And it goes all the way back to what Ms. Ellicke mentioned about, you know, our business leaders are looking at digital strategies to generate all this data. And it's all unrealized. You know, it's not... It's, you know, not... It's just hard to make that there. So it's a little scary as a practitioner to go out there and say, I'm going to put some metrics out there saying how I contribute to the value, to the use of the value of this data. And there's a lot of, you know, there are organizations that have begun to do that. There's a lot of research about that, you know, that you can take a look at how to think about value with data. But I think the critical thing from a practitioner's standpoint is to have that underpinning that says, this is the focus is delivering the value, not just following the rules, and that we need to have the... We need to have the substrate in place, the foundation in place that allows us to deliver that value no matter what the use case, no matter what kind of data it is, and no matter who the users are. And that's where that flexibility comes in. So thanks. Thank you very much for that. And thanks, Eric. I'll turn it back over to you. Okay, good. And you know, as you're talking about this, and as I'm listening to this presentation, it really dawned on me that this slide does a pretty good job of giving us a window into why we're seeing such a surge of attention around governance. It's because of the data lake. And I say that because a data warehouse is understood to be a tightly controlled construct. It's a very heavily engineered solution that was designed for specific reasons, namely getting key bits of data and insight to important people at the right time. So everybody understands that a data warehouse has a lot of engineering. It's very well thought through a data lake, however, is a fairly new concept. And a data lake, even as the name implies, is much more open-ended and therefore less controlled and less governed. And it seems to me in the silica I'll throw it over to you to answer this question that the data lake and the new importance and emphasis on data lake design is really driving a lot of the awareness around data governance. What do you think? This is Dan again. Yeah, that's certainly true. There's no question about that. I'm sure you see that as well. She's nodded. I agree. We are, you know, and I think that part of the reason the data lake is so popular is because of the fact that it gets you out of the governance street jacket that many of us have built up around our data warehouses. You know, for a whole variety of reasons, most data warehouses, you know, nothing can go into the data warehouse until it's been fully described, you know, boiled down to its smallest possible components, you know, all that kind of stuff. Whereas that's not the experience that people want right now because it just doesn't scale with the way that we're using data as we go through these digital transformations. So what we need is a process like the one that's pictured at the top here where, you know, every time the idea is a light touch and that every time someone does something to the data or gets value from the data, they also contribute to our understanding of the data and its context. Right? That's the point. So if you put it in, I want to learn a little bit about it. If you search for it, I want to know why you're searching for it. You know, all these kinds of things. As we go through that, you can build up a picture of what is the data? What does it mean? You know, what does it look like? And you know, there's a feedback mechanism here too. You can have, you know, machine learning, assistant all of that. And you can have crowd sourcing, assistant all of that. And that's really what, you know, what that modern offensive governance process and, you know, things like a data catalog will look like. Yes. And I'm going to add that it's true that there's more interest coming from data lake projects for that. But I think the reason is also because it's a lot of unstructured data, so more, it needs more governance than structured data, obviously, right? There's more needs there than anywhere else. Also, I see questions here. I don't know if I should answer it now or should come to this soon. I'm sorry. Maybe you are moderating the questions. I'm sorry. I'm not going to interview you. That's okay. That's okay. I'll throw one over to you and then if it's not the one you're thinking of, go ahead and feel free to address that. But in terms of stakeholders, I think you made a couple really good points there. And especially as you start talking about larger organizations, well, we have all these different roles and responsibilities. You talked a bit about that, analysts, you have data stewards, for example, you have business people, you have technical people who need to be part of the conversation. One of the attendees is asking, as companies rethink their approach to governance, how has this impacted organizational policies specifically with respect to inclusiveness? And I think the question there is, are organizations realizing they need to get feedback from more people from across the spectrum, not just from the top, not just from the bottom, but really from that entire hierarchy, and you need to weave all that stuff together in a way that is meaningful and makes some sense, right? So are you seeing when you go talk to clients that they now realize they must get input from each key stakeholder across that spectrum? Yes, that's a very good question. And we get that question a lot lately, that they want to define the roles of people, and they want to get the approval of multiple people. I was surprised myself. I was not expecting that, to be honest, because they want to include people in that process and make sure that everybody agrees because the experiences that all these customers, at least many of these customers that I've dealt with, they come with multiple definitions of the same thing. So they want instead to combine all this. So moving forward, everything that gets onboarded has to be approved by multiple people. So we have tools that allow you to do that through workflows with defined roles and automatically route the task to the right person and complete it fast. I think, Eric, also, the other reason that that may be occurring is because the places where we've seen the huge wins in using data to create organizational value have been when we've combined data from different silos. Data is obviously very siloed within the organization. Part of what we do with the part of what people are attempting with the data lake is to combine data from different silos. And when you do that, you wind up with diverse sets of stakeholders that need to be included. And so if we think about this in a larger picture and we say, how are we going to maximize the value of our data, that automatically implies that we're going to have a significantly increased number of stakeholders in the decisions about that data than we would have in the past where we only were capable of doing that within a particular silo. Interesting. And I ran across a very cool article just the other day that I think is worthy of comment here because it kind of is a segue to a question that just came in from an attendee. And a good buddy of mine, Tony Bear, wrote an article that basically said that cloud storage is the de facto data lake. I thought that was kind of funny. His point is that sometimes you just wind up with lots of stuff out there which you could argue is your data lake. It's just not very well governed. But there are a lot of organizations implementing data lakes now in the cloud. And one of the attendees talks about how they're going to be implementing an Azure data lake. Can you talk about how you guys deal with that kind of situation? What are your thoughts on some of these major vendors like Azure and Amazon Web Services provisioning data lakes? And what are your thoughts about this concept of cloud storage just in general morphing into a de facto data lake? You know, I know Tony as well. I think that's a very reasonable problem. It's a very, I think it's a probable outcome for many organizations. What, you know, I mean, unfortunately I'll confess my age. You know, I'm old enough to remember when data warehousing was new. You know, the idea of if we can just get, collect all the stuff and put it in one place will be better off. You know, that's an idea that's always there. You know, if you're an architect it's nirvana, right? But, excuse me, there's always entropy as well. And I think that, you know, what we're seeing with cloud storage is that's kind of the next level beyond the, you know, on-premise-to-do cluster in delivering that, you know, let's put all the stuff in one place. You know, from our perspective, you know, at the end of the day, data, the way that data is stored is only interesting because of the capabilities and limitations that are attached to it, right? I mean, information about your customers is about your customers. The fact that it's in a Hadoop cluster or it's in a relational database or it's in some kind of cloud storage, it doesn't change the fact that it's the information, it's the customer that's the focus of that. And so that's what, you know, what we need, what we try to do as an organization is really bridge the gap between that data management infrastructure that we're talking about here and all of the various kinds of tools and capabilities that we have and the increasing diversity of tools and capabilities that we have for consuming, for analyzing and for generating value, you know, actionable analysis out of the data, right? So this is, you know, the, I agree with you, I agree with Tony that cloud storage is the future of where we're going and, you know, we're perfectly comfortable working with that today from a product perspective. But I think that it's, you know, almost to a certain extent it's a separate decision, the economics of how you store and manage and retrieve the data from the economics of how you describe it and put meaning in context around it. Yeah, right. Does that make sense? Yeah, no, it sure does. And let me push this slide now and Vasiliki, I'm going to throw this one over to you. Data stewards are a role that we've come across in the last, I don't know, seven or eight years, probably seven years ago is the first time I heard the term data steward. And to me, these are the folks who are really charged with the responsibility of holding everything together because they need to speak to the business side and the technical side. They need to understand workflows and approvals and processes. They need to understand the data definitions, the competencies, as you say in your slide here, the completeness of the data. To me, that data steward is front and center in any kind of data governance initiative or role. Is that about your opinion? Yes, absolutely. I agree. And that's why in our tool also we have a special area just for data stewardship. I agree with you. I'm just trying, I think we're just trying with this slide to show the expansion of this world into additional roles. But absolutely the data steward is very central to making sure that the data is right through the workflows and the approvals and the data quality and the issue management. I agree with you. Yes. And let me understand, when you go on site and you're working with customers, how do you ascertain these kinds of things? I mean, 15, 20 years ago, approval processes and workflows were all just diagrams. Many times they just sit in a shelf somewhere. These days a lot of that stuff is baked into software and sometimes you can actually extricate that stuff. You can pull it out of the application and see what that workflow is. And obviously in the most recent past, we are seeing some really dynamic and well designed workflows that are once again baked right into systems. In fact, we have this whole information economy now that's spinning up all around us and lots of service providers doing very specific things around stuff like workflow management and decisions and approvals and so forth where before Bob would have to email Susan, who would email Jim, who would email Larry and it would kind of go around this long, arduous process. Well now we have applications that do that stuff in minutes and alert the right person. I mean that's where we're headed these days. So I guess my question is, you know, how challenging a process is it for you to show up at a client site and sort of ferret out all those workflows and really get a perspective on what they are and what they mean. Absolutely. That's a great question. Of course we have out of the box a very good number of workflows that are very much standard because based on the experience we have with our customers we built them around the standard processes of approval, of escalation, of issue management, of certification of assets. So we come out with those workflows out of the box but also the tool gives the flexibility to change. So within the tool you can change parameters, you can change roles of people who are assigned to do the tasks and I agree with you. All this makes a very big difference in the way things are done. And also your comment about emails and meetings and all this time consuming, it's going away, it's never going away completely but not for that for all these different tasks that you have to do and you can do them very fast through the workflows or even we have a comment section so you can comment on things and automatically get notified of a comment or so yes it is getting, and actually I had another customer who told me they lost millions because they didn't launch a product on time because it was not a task, it was not done on time, it was not escalated. So time is money. Yeah, quite literally. We have another really good question here from the attendee, let me get to this one, someone writes, and what about the data owner? I understand the data steward is a more hands on roll but the data owner has a place especially on regulatory definitions. This is very true. So data owner, and that's kind of an interesting concept, right, who owns the data technically, who's responsible for it, is that the DBA, is it the business person, etc. What are your thoughts on identifying who data owners are and how do you weave them into the picture? Yeah. Okay, I'll take it. I think owners, what we've shown you here from our perspective is just a subset of a lot of the roles and things that we have. This particular slide that you've got up on the screen was really just designed to illustrate what we need to do these days in terms of connecting the activities of individual stakeholders and the data to the value stakeholders, the people who get value from the use of that data, right, and do that through activities and metrics that are around that. The data owner is another role and there's a set of activities and metrics and value that that individual is going to create. Generally speaking, the biggest thing that has to do with their, you know, they are responsible for making sure that this value, that the data that is there is the data that can create value for the organization. That's not usually the way people think about it. Usually they say it's the data owner, it's mine, right? But I think we need to, you know, one of the changes that's happening is we're starting to see organizations look at that from the standpoint of the data owner isn't responsible for, you know, it's not the possession of it, it's the utilization of it that matters and therefore the owner needs to be the one who's going to take responsibility for making sure that this data is what it needs to be for all of the people who can use it, not just, you know, the people in his or her organization. And again, that goes back to some of that, you know, silo breakdown that I talked about earlier. Yeah, right. You know, one of the attendees also asked, what three basic value metrics do you recommend to align with being outcome oriented? I know you have these three metrics here. Are there any others you can share in terms of business outcomes? Yeah, these are actually kind of categories of metrics, right? You know, you probably, you'd be more specific to that. So, you know, utilization of known value, you know, that gets down to how many people are using this analysis, right? A report. A report or whatever it is, right? That tells us something about how valuable that is and how much value that's bringing to the data. One thing, utilization is a big one because, you know, with some minor regulatory exceptions, data doesn't have value unless it's used. There is some data that for regulatory purposes you have to keep around and therefore just having it around has some value. Perfect. But you've got to use it to get the value out of it. So the notion is, you know, how am I using it? What are the different ways I'm using it? How do I, you know, what new ways have I created to either use the data or new data that I've created to use in the same ways? And then there's this potential value unrealized, which is how much of the things that we keep collected, etc., are not being used and then I've got to get some estimation of how they could be. That one's a bit tricky. You know, I can't think of an organization I've talked to that I could point to and say they have a really good handle on that. But I think that's really, you know, there's a lot of potential for discovering a lot of gold in that world. Yeah, I have to think that's a pretty useful place to start, right? Because every piece of data is a potential asset and a potential liability. And certainly in some industries you do see an awareness that if data is just a liability and you're not required by some regulation to keep it, then maybe you should get rid of it, right? Absolutely. And, you know, there is, I mean, I don't know, a friend of mine, Don Laney at Gardner has done a lot of work on the valuation of data. There's a bunch of other people who have done that. You know, I think, I think this is an open bit of research at the moment that we're trying to understand how to do this correctly. But there are definitely approaches that one can take. It's probably the subject of another webinar. Yeah. You know, but I think that's the, there's no question that we all need to be kind of thinking about, you know, putting some kind of number around the value. Now, whether or not that's an economic valuation, I'm not so sure. You know, a lot of times what we're really interested in, as you pointed out, Eric, is value at risk. We have other metrics like how many issues per asset or how many issues were resolved just to be more. But a demo would show you more of that detail if you're interested. The person who is interested to see more of that. Okay, good. And before we sign off, I did want to just point out one thing. This is called linticular printing. I didn't have that word on the tip of my tongue when we gave the presentation. And you could, there are different ways of doing this too. And the whole point here again is helping you see the big picture and step outside of a particular view. And of course, to the points made by our speakers today, talking to your different stakeholders will help you achieve that perspective. And remember, perspective is and should be a fluid thing because you learn new things that's going to change the matrix of understanding that you have. And this is a work in progress. So there's never a static moment where something just hits a point and stays that way. I think we've been maybe somewhat misled by the whole concept of data persistence as we think about data and the value of data. Because as Daniel said, data doesn't have value unless it's used and if it's used then it's used in a certain context in a certain way with certain rules and policies and regulations and so forth. Actually if you do a Google search on Salvador Dolly Lincoln, the famous artist, he has a big mural of Lincoln that he did where if you're standing up close to it it just looks like a lot of interesting little images. But if you step back 15 feet suddenly you see it's a portrait of Abraham Lincoln. And you get a hint because in the lower left hand corner there's one little tile that just has Lincoln's face. The importance here once again is just perspective, being able to step back, see the big picture, understand the value of what's going on here. That's going to help you in your journey as you work through data governance. And with that I'm going to hand it back to Shannon. Can't close this out. But great job, Daniel. And Vasiliki, thank you so much and great questions, folks. We're going to pass those on to our presenters today. With that I'll hand it back to Shannon. Agreed. Thank you so much. What a great presentation. And Eric, thank you so much for being our analysts and hosts for today. Hope everyone's enjoyed the DM radio today, webinar today. Just a reminder I will be sending out a follow-up email by end of day Friday with links to the slides, the recording of the presentation as well. And thanks to all of our attendees for being so engaged in everything we do. And just hope everyone has a great day. Thanks so much. Thanks, Eric. Thanks, all. All right. Thank you.