 Hello, and welcome my name is Shannon camp and I'm the chief digital manager for day diversity. We would like to thank you for joining the latest in the monthly webinar series data architecture strategies with Donna Burbank. Today Donna will be joining or today Donna will be talking about data governance, aligning technical and business approaches. Just a couple of points to get us started. There's a large number of people that attend these sessions you will be muted during the webinar. So questions we will be collecting them by the Q&A section or if you like to tweet, we encourage you to share highlights questions by Twitter using hashtag DA strategies. And if you'd like to chat with with us or with each other we certainly encourage you to do so. Just know that zoom chief, the zoom chat and talk today defaults to just the panelists, but you may switch that to chat with everyone in the webinar to network with each other. And to access the Q&A or the chat panel you'll find those icons in the bottom middle of your screen for those features. And as always, we will send a follow up email within two business days containing links to the recording of the session and additional information requested throughout the webinar. Now let me introduce to our speaker for the series Donna Burbank Donna is a recognized industry expert in information management with over 20 years experience helping organize organizations enrich their business opportunities through data and information. She currently is the managing director of global data strategy limited where she assists organizations around the globe in driving value from their data. She has worked with dozens of Fortune 500 companies worldwide in the Americas Europe, Asia and Africa and speaks regularly at industry conferences. And with that, let me give the floor to Donna to get today's webinar started hello and welcome. Hello, thank you. Always a pleasure to do these diversity webinars. On that note, if you have not come to these webinars before and this is your first. This is a series that we hold every month. Data diversity is kind enough to record all of these and store these for perpetuity on their website. So if any of the previous topics that we did earlier in the year of interest to you, you can certainly catch them as a recording. Also, we hope you can join us next month where we talk about data architecture for digital transformation. And then there's another rich lineup for next year in 2022. So without further ado, let's talk about what we wanted to cover today, as Shannon mentioned, kind of talking about that dichotomy I guess of data governance between kind of that business approach which, you know, can consist of solutions and stewardship and kind of organizational change, as well as some of the technical data management and I always find it interesting when you're talking to someone and they mentioned governance. I've learned to kind of really clarify what do you, what do you mean, what are you talking about, and both they're probably right that great and condition, you know, everything's not everything is either or. And some people think of it as the technical data governance could be data models it could be metadata it could be technical processes it could be change management, etc, etc. Other people even that word change management. A lot of folks think right away kind of software change management or data, you know model change management. Others focus think of organizational change management, and how do we get the people really around any change in the organization like data governance, and how do we get the right stewardship roles from the business and how do we get you know the committee structure and the org structure, both are super critical, and both of them are kind of the glue the whole things together. I always think that you know the data governance lead and the data architect should be fast friends because they both kind of see two sides of that same corn. So I kind of wanted to walk through that, that hopefully some, you know, practical tips to kind of talk a little bit more about that those different aspects and how you can put them together. So if you've come to my webinars before this this framework should look familiar to you. We use a lot in our practice to really summarize our approach to data strategy which is by as our name suggests what we do a lot of and governance is always a key part of that. And where we put that is between business strategy and what the organization is really trying to achieve in the organization through data sort of as that wedge between that and then the technical aspect of data management or analytics. So you know the databases and the you know metadata management, both are valid but really I think that the people part the culture of the process is that that glue that holds those two together. And we often get questions why do you have governance and collaboration on the same line to me those seem like opposites. Probably on the contrary, I find and maybe I'm naive well probably am. But you know I don't think I've met anyone in there and all the clients I've worked with in their career that come to the work on purpose and say I want to have bad data, or I want to hide data from other people or etc etc people just trying to get their job done. And the more they don't see the value of the data they put in because they don't see where it goes downstream. So the more people understand have a common, you know principles and common goals around data, I think the rest of that comes fairly easily, because people generally are just trying to again trying to get their job done and the more they see that data is part of that job. That's where that collaboration comes in. So, more about data governance so diversity. And I work together just every year on a survey on data governance trends and I always find it interesting. And I always find it not surprising that in terms of some of the big drivers of data, or the priorities of organizations around data is data governance. And you'll see some of the stack, the stats there over 76% already have a data governance in place or planning one 86% consider data security just kind of a little brother of data governance in a way. And then, I found this one interesting over half, I identified themselves that collaboration was a big part of using a defined data architecture. And as I mentioned, I think architecture is kind of a left hand or right hand or whatever you say, update the governance they really fit together. I'm not surprised by those findings. I probably would have thought, in my experience, it's higher than 50% around that collaboration through a defined data architecture but I think there's some methods that can help that work even better. What is data governance, I'm a data architect data modeler so I always like my definitions, and I took this from the dama DMBock or the data management body of knowledge. And I like this definition because I think it really gets to some of the nuance that data governance is that exercise of authority control and shared decision making and to me that's kind of that why do you have collaboration and the definition of governance that shared decision making to me that's kind of the yin and the yang, the carrot and the stick. So think of yin and yang almost two opposites of one being very, you know, rigid and the other one being more flexible. Similar with with governments you can't only be the stick. You're going to lose interest. You really have to in all cases have that carrot to really make people interested even if you do have a big stick, and you have a regulation. And still, people may do it reluctantly you really get a better result with carrot, but you can't be all carrot without any sort of policies or stick so I do think it's a balance and you really need to look at both to make it successful. What is data governance and I'm sure I think this type of meme is maybe a little old now but I still like it of you know what what I do for a living with my friends think I do it so my mom thinks I do. I do one of those for data governance right so data governance you know might not everyone's heard the quote from Harvard Business Review that data science is the sexiest job of the 21st century. Unfortunately I've not seen that yet for data governance being the sexiest job of the 21st century story of my life. If you tell someone you're in data governance, you know what are your friends that okay some nerdy data thing up in your, your, your office somewhere and we never see you just sit in your computer, probably not too far from the truth. Right what my mom thinks I do or my dad. Well, data is kind of like a librarian and in governance you're kind of telling people to do things all the time shush be quiet do the right thing with the books back. Again, maybe not the sexiest of stereotypes there. No offense to any librarians in the call. I think librarians are awesome but again, not that what society thinks I do. Oh data governance that must be like data science because I've heard that's really sexy but you're like one of those crazy data scientists that do crazy scientific experiments on data or something I don't know data, something what my coworkers think I do. And this one be careful of the other people that are always telling us what to do, you're always yelling at us you're always saying I can't. No, no, yes do this follow the standard right. And although again back to that carrot and stick you do have to have some policies in place but again that you don't want to overdo it with the stick that everyone's afraid of you or everyone hates you or they see you as a terrorist that's probably not the reputation you want to have in the organization. What I think I do. Hey I'm saving the world through data I am the governor tour. I didn't even know this existed. We actually had a client that wanted to name her head of data governance, the governor tour because she thought that would be a lot more exciting in the in the ad. She's still down she still called the person the governor when they got the job but yeah I do think we often do feel like hey we're trying to save the world here we're trying to save the organization, and sometimes we feel like the lone superhero, you know fighting this battle on our own, but probably the reality of all this what I like to say and I think is true. What you're actually doing is driving the success of the business, most companies at some aspect are data driven. Something else finance departments data driven, and they want to make sure that their numbers are right but more and more companies are really realizing that at their core, they're a data company, or data can help them be a better whatever kind of company they are through better efficiency quality, etc. So the more you can align what you're doing a data governance with that success of the business. That's the biggest carry you can get right. And so I think it's important to remember that and then we won't dive into all of this in this presentation, but try to get more ROI and metrics around that to really prove to folks that yes it does matter if you have the right product code on this product because we want to know how many we sold, we want to ship it to the right place, and making some of those, you know connections with either dollars and cents or live saved or whatever your, your organization is. I think that's really the crux of what you try to do with governance to get that buying. Okay, so what is in terms of that need or establishing the why behind data governance because I would say more than half the battle is aligning with that why getting everybody understanding why this is important if they don't understand it already. And then if they understand at a theoretical level that it's important understanding that no we mean you like you have to do things to. So it's very core, if you think there's kind of two half sides of that coin of data governance you're either reducing risk, or defense, or increasing opportunity offense. So, I think with the reducing risk, probably not a lot of surprise there again I think we probably overdo that sometimes with governance yes there's a regulation yes there's GDPR. Yes, there's, you know, pipeta or, you know, FERPA, or be HIPAA, or, you know, name your regulation there right. And that even if without a regulation there are certain actions or decisions that you know there's accountability for the information you're doing you work with a restaurant chain for them. So if we have the wrong recipe or ingredients on our menus and someone gets sick, you know, a reliable for, you know, the moral responsibility of someone's life or and or we get sued and or we end up in the newspaper. You know so it isn't always a regulation for that that risk aspect. But what we don't I think think enough of with data governance is that increasing opportunity that data is a strategic asset and we're going to be much more competitive in the organization. I would say in my experience, having done this forever. I see this more and more now, I think you don't have to convince as much as we did in the, the good old days that I think more organizations see that we want to be efficient, and we want to be competitive and data helps us do that whether it's product data customer data, student data, you know, patient data, etc. So if you're a data driven organization, if you've been on some of my past webinars we talked about that a lot kind of that business value of data. That's, that's a no brainer to use that phrase to kind of connect the value of data with the value of the business when you're a data driven business. Some of the other areas that are kind of aligned this that approving efficiency, kind of talked a lot about that one already, but not only does it help you run the business per se more efficiently, but just the data is a lot more efficient I mean the number of times we come in and do a data strategy or an assessment. And maybe the data quality is great for the organization. But some poor person that spends weeks every month fixing it or every time we have to do the financial reports for management it takes us, you know, five days out of a month to get that done. That's not efficient so you know even the company itself can be more efficient but the data centric roles can be more efficient. And this idea of driving collaboration and accountability. We use those words a lot. So maybe because I'm sorry but data governance isn't always the sexiest phrase it is becoming more sexy. But the idea of taking quick collaborate around data or can we be held accountable for data, you know if you, if you lost some money, you'd be held accountable for the organization if you lost some product or shipped it and it got damaged, you're really having that same if data is an asset, you are accountable to the same, you know, what do you call it, the same level of accountability. I've joked a lot already that data isn't sexy but data governance isn't sexy but actually it is becoming it's funny. You know there's all these buzzwords in the industry big data and data science and all that. There's a lot of action data governance at one, one company, and one of the younger younger kids I'll say he's like yeah I know you love to throw those sexy buzzwords around like governance and everything but we said wow that now things have changed. I didn't know we were the cool kids, but we really are and I think you know we get a lot of our projects come in are really driven by things like data governance and data quality. On this note of offense and defense. It's good to think of when you're designing your data governance organization and or selling data governance which you will always have to do continually over time or do an organizational change management effort of what is the driver for governance and getting this wrong can really have a negative effect. And so, you know, if you're, you're a startup, and you're all about profitability and growth and revenue and competitive advantage and you are the one to come in the room and say whoa whoa whoa. Let's just wait a minute you know there's GDPR and let's get all our metadata in order. You might be right, but the tone of that, you're kind of the downer in the room and they're going to stop inviting you to the meetings by the same token. I probably tell by the way I talk I tend to be more of an offense opportunity driven kind of person, and I've made the mistake of going into a very risk focused insurance company that maybe just had an audit. And they are all about compliance and regulation and avoiding audit. And I talk about all the sexy things you can do with data and the offense, you know you don't want to do go that far, either, or, you know you're talking about patient or student data and you know, I think I made this mistake, not that I ever make mistakes, and you talk about data monetization. You know it sounds like you're going to be selling patient data or selling student data. You don't want to be doing that as well so you know read the room, read the company, and kind of thank you on the spectrum and my company are we more driven by the offense and the opportunity, and or more driven by the defense and and probably you're somewhere less with that purple in the middle is you know you're probably not fully read or fully blue, probably some purple you know, even if you are very compliance focused, you know company actually we work with one of these was a very compliant focused company they literally had been in the paper and they literally had had a huge fine, but the way we still we sold data governance was through opportunity, and how they could have better customer satisfaction and all that, while they did the compliance and regulation kind of put a little icing on the otherwise not not pleasant opportunity. Okay, so I talked in the beginning about you know you can say the word data governance and people can have vastly different things one could be talking about again metadata lineage, and the other person could be talking about, you know committee meetings. One of the issues of that is there's something for everyone in this right or something for me to hate to kind of depend so if you are the I often get the question in my consulting is you know who should be leading data governance, who should be that data governance lead, and I fairly strongly say that that's a business centric person someone that can you know is enough about technology that they can be aligned with it, but it's not a tech person that should be someone from the business or even project management that loves working with people that can be an evangelist that can read the room and understand motivations and why people would want to comply or not comply or etc. There's another side of data governance that really does get excited about metadata lineage and the standards and aligning data types, you know, sometimes you can get that perfect person who does both. It's really a spectrum, and they both are necessary and both of those need to work together. I often see that as the head of data governance and the head of data architecture, the data architect, and they should really work closely together and augment each other, but probably important to remember, because you probably are doing part of this in your job as well. Okay, when we look at data governance we like to apply standard framework. This is ours. And as I kind of have been talking about already, you know key to that is getting that vision and strategies for the organization, and then how you can align with that strategic vision through data, you know why is this important to people. You know is it to improve revenue, is it to reduce risk is it to, etc, etc. And the foundation of that is the culture and communication. I would almost say that a good data governance rollout plan is akin to if not aligned with an organizational change management program. And if you're not doing that, or perhaps even your marketing program, but more and more companies we work with that are really trying to do a data transformation or have a big data governance program. You can work with a change management group, or use some principles of change management ad car or some of the other methodology surrounds change management that really help drive behavior and change in the organization. And then so if the vision and is the kind of guiding light and the culture and communications kind of foundation to make that happen. And if the pillars of that are the sandwich in between are the organization of people what are the roles how do you organize them is that a top down hierarchy is it federated. We'll talk more about that. What are the processes and workflows and I think there's two sides of that what one would be. What are the data governance processes, you know how do you log an issue how do you do change data quality remediation, etc. But some of it is how do we align with business process, especially if you're doing something like master data management. I say that's almost 60% business process and how you align the business processes to have people even enter the data correctly from the get go data management measures again. This is almost that flip side of the the one I showed here right if on the left is the people and on the right is the tech shouldn't have made the tech the devil. No offense, I'm probably the person on the right so no offense there. This is sort of mirroring that right the right is more gets more technical as you go. So could be data models it could be metadata could be lineage, etc. There are tools and platforms that could be gosh there are tools that sell themselves with data governance tools it could be a data model. It could be a change management tool, it could be a process model. You know I often say please look at that last and we'll talk more about that number of people who come to me, they say I'm doing data governance what tool should I buy. And I say that's the wrong question, you may need a tool but please don't start there. I had one client and one of my favorite client we really got along we still keep it to us it was years ago. And he came to me and he said I want to buy tool X my name the tool I don't do that don't make me ask. So and I said, you know Mike you don't want to buy that tool and you shouldn't start there is like but I want to buy tool X and we did a whole data strategy and data in like eight months later he comes back like can I buy tool X now. You didn't just say that he said it with a twinkle in his eye so I think he knew he was wrong but I still think as soon as I left the room he'd want to go buy that tool. It's not the way to look at it there are tools to be helpful but don't drive it from the tool, the vendors will kind of push you in that direction and I used to be a vendor so I can tease them as well. But be careful it's there are some helpful tools you may not need a tool your tool might be PowerPoint to help sell this. But there are some of the excellent tools in the market might be the wrong tool for you in your use case. You know do you need to have a tool to help with workflow or is it more metadata management and, and I've seen a lot of companies that we've worked with that buy a really fancy tool, and it just doesn't work for them and again it's not the tool it's just they didn't, they didn't think through this whole framework before they looked at what tools they might want to need. So I'm going to read through every row and column of this in great detail. I'm kidding, but I thought this might be helpful of what types of things and are we talking about in each one of these, you know, aspects of this framework might be a good reference if you download the slides later, you know what do I mean by organization of people who not only are in your data governance committee, but maybe who's the stakeholders we need to work with either inside or outside the organization. Or regulators, etc, etc. So I'll kind of leave this for you to read through, maybe can jog some ideas maybe areas you hadn't thought of in your framework. You know, we do a lot of data governance frameworks and implementations and it's probably half and half of people who haven't done one or starting fresh, or people who have started one, and it just isn't singing like they wanted to. So how can we tweak that maybe we hadn't thought of is it our committee structures wrong we have wrong people in the committee we're doing too many committees and not enough action like what is it. So maybe some of these questions might help you job for some ideas for your data governance. So in terms of who the people is a huge part of governance no matter how you slice it. So when you think of who drives data management in an organization this is from one of the data management surveys, I did with diversity. You'll see that probably not surprisingly, there's the roles like it manager chief information officer data architect. But what I find interesting is more and more as we do the study is either the data governance lead or a chief data officer or business stakeholders that are involved in or actually running the data management program so that that's a good and bad to have the if you say the people who are running it, they should be right I should be driving it, but they probably shouldn't be building your analytics platform right so where do those roles fit. And it should be driving it but they shouldn't be driving it in a vacuum without understanding the business needs etc so how do you get that right balance and how do you set up again the org structure around governance to manage that so one area of data governance is data stewardship, I'm using data stewardship in a more general term there are often a role called a data steward. There's also other roles like a data owner, executive champion with technical data steward, people get more and more creative data custodians data. I don't know I've heard a million things right whatever makes sense for your organization, but data steward ship and generalists you've got the business people who don't have data in their day job title, but are responsible for the quality and the accountability of that data and how it's used. I often find so we often we come in and do a assessment or something we talk to a bunch of people we do a bunch of interviews, which I think is sometimes a luxury that maybe working in an organization. I have, but I would say maybe don't think that I've had some. I know some people who have had very successful data governance and then they've worked in the company and they do that they do kind of a listening session they go around the org and they, they ask people's opinions on what's working what isn't. And you're doing two things one you're finding the challenges in the organization. We are often kind of finding these hidden heroes, almost every engagement we've done there's, there's someone in the org that we bring up later to management and say you know, Mary would be a great, you know, Stuart and Mary. Like, did you know that she had this spreadsheet this basically your vendor master data that no one's ever paying attention to really money should be a great advocate. There's always someone like that or many people like that, that no one's ever really asked, or, or they want to be involved and made them in vocal but they don't know how or they don't have a structure around it so I'm fairly confident, you will be pleasantly surprised that there's more people than you think that are probably on your side, and are, maybe they don't know how to verbalize it or don't know what that role is but they know they're having problems with data. They all have been creative and trying to solve it themselves, but the more you can, you know, create a framework around that so people are, you know, paddling in the right direction together. That that can be really rewarding not just to you but to everyone else who's involved like a lot of people are very relieved to finally be a day of Stuart because I can fix the things, things they're working with we work with a big financial institution last year, and one moment said this is my dream that I am finally before I retire, able to fix the things that have been driving me crazy for 25 years. And maybe that was a little extreme but she was thrilled to be a day of Stuart and that was almost her career goal is to finally fix this stuff. So yeah, listen to people you may you may find some hidden friends. In terms of roles, you know the day to management body of knowledge to embark is a good, good guide. We, we pay a lot of attention though are these the right roles. How many levels of roles and what the names of those roles should be, but in general, the, these types of roles generally often exist. So, generally, you should have some sort of executive sponsor that should be a non I wouldn't say it's your, you know, head of it, it should be your head of marketing your head of finance or someone or the CEO, whoever you can get on the executive team that that really is a champion for data and it's going to help drive this from the top down. There's some sort of business data owner who they are representing the data for their area or their process or, again I could do a whole webinar and how you define these roles but they're looking a little more more big picture. Are there general business roles and policies around this data, you know how do we manage it do we share it what are the key KPIs or measures we want to track for the organization and that sort of thing. The business data steward is a similar role other a little bit more on the day to day, they might be defining the more detailed rules around the metrics or or the data types or the, and you'll be surprised, you know, they might be there arguing for hours over what the right data type is for a value or, you know, the detailed calculations of a KPI or something and they often work very closely with the business data owner sometimes report to the business data owner, but are more the, you know, the hands on the feet on hands and the partner, but they're actually using it day to day and they're probably a little more into the weeds. I do think of that carefully I've seen, I've seen that kind of go wrong on both sides, either you have the business data owner doing way too detailed stuff, you know, hello Vice President of Operations could you just go through this spreadsheet and make sure all the calculations are right they really shouldn't be at that level and they'll probably bore them and they're too busy and you got to turn them off at the same token often you might be trying to get that bigger picture vision from the business data steward, and they, they literally almost can't see it, they're so stuck but they're so involved in that day to day and in their process, they just haven't had the opportunity or don't have that exposure to the bigger picture. It'll be great to tell you how that system works or how they do their job, which is what you need, but they probably can't see the full organization because that's not their role. So be careful about that. And then there's a technical data steward, often, and they either are the subject matter expert for a system like they run the CRM or the, you know, point of sale system or something. I just be careful there. Some people associate those folks with a business data owner, we're just going to, you know, manage the CRM and think you don't don't think about systems. You should be thinking at a business level, right. So these people are important, but they should be aligned with the business data earn beta steward to support them, but they shouldn't necessarily be driving. They should be supporting. Okay, so about organization and change and capability and I think this is another one to think very carefully about. And again, where we see things go wrong, this is often it so look at your the one on the left is more your organizational structure and capabilities and how your company runs. And the one on the right is how you might create and define a data governance organization that you have a standalone data governance organization. So the one on the right is almost your classic dam a DMBOP type thing. This is genericized but you have some sort of executive leadership team with a sponsor, you would have a data governance steering committee with your data owners, maybe a data governance committee with your stewards of working groups, etc. That can work really well we often have structures like that for some companies though that just seems really hierarchical and really complicated and too much. So if you have only one committee, maybe you have no committees and just roles to start, maybe you have a more federated approach, maybe you integrate this with the agile life cycle. We had one company that actually integrated data governance as part of their product release they were a retail company, and every product stage of the product lifestyle product like the actual widget, they were selling. How to data component is the data right for the shipment is the data right for marketing etc. So, in a way that's the best data governance you can have because it's operational is actually literally part of people's day jobs. So give that a lot of thought that is often where it goes wrong, people come to a committee and they, they hate committees, and that's, that's just not the culture, or we don't have a committee and it's a very committee driven culture and we don't have the right people in the room to make the decisions right neither neither is right or wrong it just has to fit your culture a lot of great things about committees just has to fit the way you do it as an example. This is, I use this one a lot because it's kind of ironic to me or interesting to me. This was from a manufacturing company that we initially started with something like the one here where it's very hierarchical and committees and they said, you know, we're not a hierarchical company. I would have thought with manufacturing it was just very XYZ and, you know, maybe a top down wouldn't wouldn't be bad and having that structure would be good. I've actually changed this a little bit because they said you know we want not to be stereotypical, but these were all kind of guys guys kind of it was a manufacturing they had factory floors and all that big boots and you know, and they said, you know we want more on the presentation concentric overlapping circles and pastel colors, because that's more soft and we, we think ourselves as overlapping teams not boxes and lines and hierarchical structure, and we're more agile or more fluid, which again would not be the stereotype I would have had for you know, manufacturing for big heavy metal things. But that was their culture and I we listened because it wouldn't have worked to have that rigid structure, and it more came out like something on the left, where they had a council, and then each group kind of had their own autonomous voice but then they got together as more of a, you know, a federated collective and then the actual action happened in their agile project teams. So that worked really well for them, but you had to really listen closely to the organization so you wouldn't. And if it had worked had we had a different structure. And I would say, I just talked a lot about that but it is important and that's often where we see things go wrong. You're just not fitting the culture of the company. So, how do you find the right balance between kind of that business approach, the technical approach, and the other part is kind of being more proactive or reactive and this might be a complicated slide but I like it. I can read it the one is, are we more proactive which means resolve it at the source I mean if you think of data entry if all data quality. If all data systems had drop down menus for the, you know the states you can enter the codes you can enter and, and then the right value check your data quality be great, you wouldn't probably even think about it. So, you know that would be kind of your resolving it at the source, resolving it after the fact is when you're doing data quality cleanup. And that's what you probably don't want to do. So what can you do upfront and I have this both automated things at the bottom like what can we have, you know, data quality validation at the source when you put data in. Can we have automated validation checks, etc, we've automated workflow of the top in the business do we have business process change to really make sure data gets in right with right policies procedures, do we have the right data governance is even setting those right policies right so that's really kind of the thinking ahead. So you might be doing things in the back that are kind of that reactive we're cleaning up data quality or we're, you know, doing transformations to make the reporting right etc. And then I put that one in the middle, which is kind of somewhere in between, it could be you know the data stewardship over time auditing quality over time. That might surprise you with the conscious disregard right sometimes you actually Shannon and I were talking about that before the break of, we tend to be data quality people and, you know, you could have a spreadsheet of data of your customers and could be 100 customers and you spend all day trying to get the data just perfect you know the person put their address is 101 Main Street should I, should I write the whole word street or just put ST or you know it might not matter we're just, you know, I don't know we're going to send Christmas cards to the middle it'll get there don't worry about it right. So, or maybe this data, you know we had one customer that outsourced some of their data quality and the results came back and said, you know your fax number fields are terrible, you know half of them are empty. Well, who uses the facts anymore I hope nobody because I hate them, but that might be a great conscious disregard yep fax numbers empty because we don't use fax number anymore what don't worry about it move on. And so kind of that prioritization of what to worry about the super important as well. Okay, so on my tool rant. There is no one size fits all data governance tool no matter what the vendor says they may have be a data governance tool, but there's many types of tools that do data governance so this is trying to show. This is actually from one engagement we several engagements we've used this kind of pattern. This is one example you don't have to do it this way, but one would be like what are the functionalities you're looking for, we're trying to do processes and workflow is that a technical lead is that how important that is to the business. And then kind of look at your tool so you can do different visualizations you know for the one on the right. Their most important thing was glossary and metadata management, your security wasn't important lineage wasn't as important to them. So they, they could, you know, maybe even have a SharePoint list of glossary terms for them maybe that's enough. Maybe your ranking would have been very different you know it's actually about security and privacy and the lineage of PII across the organization and maybe you need a detailed tool to do that. You could say hey we want a tool that helps show the roles and responsibilities and workflows for approval. That might be another tool so again all you could it might not be a could be a data modeling tool could be a master data management tool could be all of the above right so I think the best way is to start at what what functionality do we need, and then do we even need a tool for it. It's going to be I'm creating a data governance organization I can do that PowerPoint yeah I guess that's a tool but you know I probably don't need to go purchase said something. So just give that some thought. I do think data architecture it does have a massive part of data governance. One thing we often do and often get questions until people see it and then it just works really well is to have kind of pilots or proofs of concept or whatever but around architecture data governance and that often surprises people because often a POC or a quick win is oh just put together some crappy piece of software that really shouldn't go to production but then we are going to put it in production and then we'll complain three years from now and it's not working and the data quality is terrible and doesn't meet our needs right it's just often fast means, you know, do it badly and slap it together. We've had a lot of success with do an architecture POC just pick a business problem that needs to be solved. Do a small data model for it could be five entities or you know whatever to really understand this one I think insurance company looks like right so that you know I have a customer it makes a claim. I have a policy that's linked to the customer not the broker or even something very simple like that can actually often unleash really interesting business rules and root cause analysis. Do a process model how does data get into the system and who touches it along the way a good old fashioned crud matrix of where data is created read updated and believe it can go a long way and really highlighting a problem. Do an architecture diagram around the systems that are involved in this, what are the business rules what are the policies what's the put a glossary together. Is there let's do some data quality profiling right so kind of pick all of the things you should do in a data governance initiative, get your data stewards get your data owners together and just do it with a particular question you know how are we going to support our brokers with data how are we going to support our customers how do we price our policies. We're trying to price our policies but we can't get credit rating, and that's going to be a problem. How do we, you know, anonymize credit rating so we can, you know, rate on it, etc, etc, but pick that problem and then again sometimes you might have problems getting your data stewards to get excited about it, and until they do it. And so I often kind of learned by doing or, or sneak it in. I don't know putting chocolate on something so your kid will eat it right it's, it's, you know, don't worry about this is just a quick, you know thing we want you to do. And then you turn around and say wasn't that helpful wasn't that great and they agree and you say that's data governance. Oh, well if that's what you want me to do I'll do that again just don't make me do that data governance thing right because I have visions of big boring committees and things like that but once you actually do it and they see the value and see that really it's stuff they already know in their heads, they're just getting the right people in the room to solve it. You'll get the buy in and you're starting to do the right thing along the way. If you do a lot of these. Gee, by the end of it we have an enterprise data model. It's a process model for the or, but if you went to management perhaps and said can we have, you know, a year and a half to document all our business processes to improve data quality. I bet you'd have trouble getting funding right, but if you say a series of, hey we want to clean up customer data so we can sell more policies, just give us three months, sure. And then you keep doing that and showing the value, then maybe you asked for the big chunk, but often you can kind of do it by stealth through these small targeted projects that are still doing it the right way. Okay, so it's a little bit more into some of these tools. I'm a huge fan nerdy fan of data governance and metadata management and metadata is often I like to say that's the kind of the actionable aspect of your policies and procedures. It's a way that we need to a non, you know, we can't share PI, or we need to have certain formats or we need to do whatever it's the metadata in the systems that actually allow you to do that, and make it, you know, seamless and integrated to actually, you know, enforce these policies something like GDPR how do you, how do you know that where your customer data is really can't do that without metadata so you can have a great policy say you know if someone asked for their data you need to share it can't really do that without great metadata. There's a lot of tools on the market that support metadata can do a whole webinar just this one so I'll try to keep it simple. But just as some general categories I know there's a lot of ways to manage metadata, kind of the biggest, biggest baddest if you have you know the need and you have the budget and would be kind of some sort of metadata repository catalog data catalog is kind of a new sexy word. And at its core is really almost a data warehouse for metadata, where you have a metadata storage of what they call metamodels that's almost like your data model for your meta you what are you storing your mic storing information about you know tables and columns and etc. There's some sort of matching and reuse logic so that I know that a customer entity and my data modeling tools the same one in the database and kind of linking that together, generally some reporting some sort of portal or integration and in some way to integrate and get that data in whether they call them scanners or interfaces or every vendor has their own thing, but some way to get it in and some way to get it out so it's actually pretty simple, but there's a lot of complexity and actually we've done it at webinar earlier in the year on metadata catalog repository and how to choose one. There's a lot of new ones when you go to buy one but at its simplest core, it has some components to look for of storing the data, getting the data from the different sources rationalizing and then publishing it to people. Okay, so again if you have the need a huge fan of those I used to build them and write them and sell them. So, you know, more standing bad about them, but I will also say, you can often get away with a specific tool repository or functionality. And yes, I have, I have written and sold and developed metadata repositories and I will also say, you can probably get away with a SharePoint list for your business glossary to start without spending a whole and both are true, depending on your use case right. I may not end there, but it might be a fine place to start. A lot of the data modeling tools when you think of it a lot of your metadata is already in your data modeling tool your entities and if you've done it well, your definitions and some of your lineage, and more and more of the data modeling tools. So utilizing that and are kind of putting a data catalog layer on top of their tools. It could be a data dictionary, some of the database vendors again are realizing this and putting a layer on their tools and data dictionary. ETL right there, that's a lot of your lineage. Again, a lot of the vendors are putting a layer on top, so you can kind of see bi tools have their semantic layer. So you could, depending on your use case if you really are just all your metadata is in the data modeling tool. So that's enough to do that. If you're really just more focused on the database development, maybe a layer on top of that is fine. And again, I'm a big fan of and and not or a lot of these tools can be integrated with the metadata catalogs. So you could start with a data modeling tool, and then as you expand other sources you might want to scan that in and have that in a bigger data catalog. And I also don't want to diminish this idea of a metadata exchange or registry or, again, when you're thinking of open data, if you're trying to share data with any other external party. You want to have that in some sort of document either structurally JSON XML, and or open data publication with, you know, definitions of the terms and things like that. So, some are all of those approaches can be really helpful. I think the models of talk a lot about big, big fan, you can either do it at the business level, and or the technical level. Again, we have a whole webinar on this but I would just say be careful of knowing which level you're talking about right so if you're at the enterprise conceptual logical, a big part of that is the business rules. If you're at the logical last physical, that's going to be your technical data structures and your data types and your naming standards and things like that. And those are super important to governance right so if you're trying to, you know, develop correct reports you need those conceptual and logical models to understand even what these ended what is a customer, what is a policy how do we define the rules. And at the physical level, you need to make sure all those standards are enforced you have the lineage etc. So it's a lot of the tools out there, especially the more expensive ones can can do all of those layers within the tool, and you get the lineage between the tools well. As I talked a little bit about this before, but whether again whether you need a fancy tool or you're doing it on a whiteboard or a visio or I guess it was a tool but a lighter tool. But again, showing those women showing the who who's using the data where is it touched that this, I'm a, this can often highlight some of the best data governance and data quality issues because you're literally embedding it within your business and you don't even think about it, because when I developed a code name for the product it therefore integrates with MDM and it's automated because we have those rules in there and then you need less that remember that slide with the proactive and reactive, you need a whole lot less reactive. When you've been proactive and it all just goes smoothly I've worked with a handful of companies in my career that I actually love this when you sort of ask some of these questions of, you know, do you have this problem with data quality and there's no, I mean it just comes from the warehouse or no we have a process, really there's people out there that you guys don't realize how smoothly it runs right like all of us you don't realize you're not sick when you're not sick. But it said then we get the cold you realize how good you felt. But that's it. Unfortunately, with data governance your reward is that you're sort of nothing happens you're ignored and everything's boring everything just smoothly, which is good. Now, but you sometimes have to remind people that there's a lot of work to make it good. I think kind of in a way the new hip version of process models is the customer journey map, maybe that's jaded but in a lot of ways it's looking at a process of sorts is the journey a customer takes along their process to interact with you and this can be another great way to do the customer journey map or work with marketing who has these customer journey maps, and then do a data overlay. So what what data are we collecting from the customer at each point in time, or what data does that customer need to see at each point in time and then who's governing that. And one question I often get is, you know how do you, again it could be a whole webinar but you know how do you organize data stewardship shouldn't we have like a customer data steward or a product is steward. I really push against that because here's a great example who owns customer data. Well, the marketing team and the sales team and the support team, and the, you know, etc, etc. You really need a cross functional view you might have data architect who's looking just a customer, you should, or master data management person just looking at customer but really when you're creating the rules, you should look across that whole customer journey student journey patient journey, you know whatever you're, you're developing. I'm not going to mention crud matrix or terrible name for a wonderful tool could be drunk or whoever comes up with a better name that we get used with those acronyms. I'll power to you. I will use that word. But what that does is take something like a customer journey or a process model and just be a lot more specific. You know this particular piece of information, where is it created that's often your owner, then where is it updated, deleted, you know, and you can often find a lot of issues that created six time we're working with a big financial institution now and we did one of these, and it was created six different times in different systems, everyone thought they were creating it. But that obviously leaves the problem so a lot of something simple like this can really highlight a lot of these data governance in terms of who owns it, and then process and data quality issues. One quick story before I wrap it up because I see there's a lot of questions coming in or at least chats coming in. This is, I think one of our success stories that did use that kind of quick win data architecture and data governance by stealth, and it was super successful so this was a big retail organization that both manufactured shipped and sold the product so they kind of had, you know, vertical horizontal coverage, and they had a lot of problems, just even tracking their customers and tracking their products. So you can imagine with customers for example, they would come into the store and they'd talk to a sales rep, and you'd ask for you know what's your email and they'd say you know go away at leave me alone.com or something like that. Then when they needed to have support later, or you know get their product shipped. Now they weren't getting notifications because they had a different email that, etc that everything wasn't linked in so very basic stuff but also very impactful. So what we did there was pick one of those really small we did a data architecture data governance spring where we built process models we did data flow diagrams we did data model, we did a system architecture diagram. We created data stewards and a small data governance framework, and then we took it was just a month, we did all of that and we highlighted the problem of why they were duplicate emails, what it was affected. And then we had all of the business people in the room. The result of that two quotes I loved. One was ahead of marketing where she said you know I never really feisty brilliant woman but not technical. And she said you know I never thought I'd use the word data flow diagram in my life, but she I'm printing it and putting it on my wall, because no one's ever explained my marketing campaigns didn't work. There was actually a kind of after this they actually had all the things literally printed on the wall, and when everyone needed to change something they would go to that wall and look at it. The other one I love was the head of sales, who actually said, huh, I never understood where that data went shouldn't I govern my team to actually use the word govern to maybe put in the right email and other guardrails we could use. I never thought that the head of sales would say please could I have some better data governance for my team. But he saw the result that was the key thing we did just enough architecture just enough governance that we told the story, and then got the buy in and we had all the right artifacts we did all the right things, but it was just having those small chunks and getting the right people in the room and solving a problem and we went and solved another problem. So, in summary and then I will open up for Shannon for questions. When we talk about governance is that orchestration of that people process technology and culture, and those are a lot of things to manage and that's why data governance can be challenging. When when you get it right like this example it just sometimes seems so easy. So that's worth the effort we solved the problem got everyone bought in, but there's a lot of like a figure skater right a lot of work to make it look easy. And there's no one ties with Saul so you know I could have gone in had that success and then gone into another big financial institute or healthcare company and tried to do the same thing and that might not have worked for them right you need to listen and do a right size call for your solution, but at the same time, no matter who you are, I do think you need some sort of quick win, because humans are humans and we get bored, and we need to see positive reinforcement or we'll go on to something else. Again, if you enjoyed these webinars that they are recorded in the past and please do join us next month for data architects for digital transformation. If you're interested in any of this shameless marketing plug. We do this for a living and would be happy to help. And with that, I will open it up to Shannon for questions. Donna thank you so much as always for another fantastic presentation if you have questions for Donnie put them in the Q&A portion of your screen by the Q&A panel in the bottom of your screen for that feature and just answer the most commonly asked one is just a reminder I will send a follow up email by end of day Monday for this webinar with links to the slides and links to the recording for everybody and anything else requested. And don't I got to say I learned another another, you don't like fax machines, another court. I don't like. So, I've been here, what are some key items that you would recommend that are included in an enterprise data governance policy of key items that are included in a policy. One is getting the definition of what a policy is what a what a policy is for one company is a procedure somewhere else and getting some examples in your organization again I often come in as a consultant so this may be obvious to you. But get some examples of other policies. I think it should be who would be impacted by that some very specific, you know actions for that policy and some guardrails, even just some simple things of when that policy was created how long it's been forced. Yeah, those are some things I would think through. I love it. And what are the key points when selecting a tool for monitoring data quality of the whole organization. I got the tool question. Do you need a tool for one. Could you, or maybe start with some proofs of concept with some tool. Can you do some simple SQL queries and write your own little dashboard that might be one. It might be, you know, how much you can see that one of the reasons to do that is kind of a proof of concept of what are your business rules and how complex they are so often there's a kind of a business rule that they have that how much you can customize that versus out of the box and which one of those is important you might just say hey you know we've addressed data. That's perfect example address data. I'm sure people have done addresses before I don't want to think through this at all. Just could you, you know, clean my data could you even augment data do they have third party tools to help augment that, and report on it really quickly. And then another aspect of that might be can you integrate with real time systems can you be that you know I said data entry at its core. How do we integrate with that. And you might just want something totally automated please I know this can't be rocket science other people have done it. And some of those aspects that I mentioned the automation the customization of business rules and pushback to systems the reporting. But you could say and we've got clients that do that actually our customer base is really unique and we have international addresses and names that don't conform to the typical naming structures and we want to customize our business rules. So that might be, again, almost your classic example, where you might have very different tools. The address isn't important to us we do everything with email now. And so I don't even care about address so anyway, those will be some of the aspects to look for in a tool, and there's some really high end you know Tesla type of tools and sometimes you can go really low end so don't don't overspend also usability. We had one customer that got a tool that had everything they needed but they needed so much training. They just didn't use it and so a lot of those tools have come a long way with being much more user friendly. Or, or could it be part of another tool like for example MDM or master name is something I still have, you know, it doesn't have to be only a pure play data quality solution that could be integrated into another tool set of data catalog or MDM or something. Yeah, you know the recent add-ons to that question as well. We created a data governance framework interchangeable with a data governance strategy with respect to this. And another person commented on that same question, you know we created a data governance module and our enterprise governance risk and compliance tool it's been very well received because you can view the data from the process business unit model or application perspective. Yep, no that sounds good. And I would say that question strategy framework. Yeah, I think I mean I want to see what's in it I mean I think why we say framework or even strategy is are you looking hope before you go and execute anything or define your data stewards or do any of your data cleansing or anything do you have kind of that framework in place or the structure of it and how do we have the policies and how do we execute them and all that before you start running so whatever you want to call it. I don't care so much I think it's kind of what what you have in it and that that's where we can have that house thing we look at at least you have the checks in the box of how we kind of looked at these things. And if yes, you know, call it whatever you wish. I think we have time for at least one more question here. The many different templates for process data and customer journey mapping are useful but are there recommendations around which to use for different types of organizations or should multiple templates be used to provide all perspectives and if so is there a best order to complete them in. I'm a big fan of taking the standards and then tweaking them to meet your own purposes so the BPM man or kind of your swim lanes. This one this type of one I think it's really good for manufacturing, or if you're very process centric like they actually this I think was some man, you know they built their product and you do X you do why you do Z and this really clear swim lanes I think these work really well. This actually is almost classic BPM and you know your classic swim lane workflow where this workflow. This type, I think it works really well because most people can understand it even if you don't know that that circle is a start, you know entity and the, if it's filled in it's the end, you kind of get it. I, for example, I put a little picture of a person here on mine because that gets to know that those are the swim lanes or a picture of a database versus a, I'll put in a picture of a file or you know spreadsheet of telephone if they're making the call so I tend to kind of spice it up. It depends on your organization, but I think these are often what it is very, you know, when the issue is passing data between groups, or it's very process centric I think these work well. Someone who does customer journey maps might cringe at this one that's just super high level. There's whole methodologies and customer journey mapping, but I think it could be the same company right this could be done by marketing this could be done by marketing. So I think they're, I would see them as separate things. And yeah, just you this would be probably looking at more from the, you know, the stakeholder point of view like a patient or a student or a customer. If that helps. But that does bring us to the top of the hour here. Donna again thank you so much for another great presentation. Thanks to all of our attendees for being so engaged in everything we do. And all the great questions and comments. Just to again a reminder I will send a follow up email by end of day Monday with links to the slides recording past webinars and upcoming webinars for this series as well. Hope everyone has a great day. Thanks everybody. Thanks Donna. Thank you.