 Hello and welcome. My name is Shannon Kemp and I'm the Chief Digital Manager for Data Diversity. Thank you for joining the latest in the monthly webinar series, Data Architecture Strategies with Donna Burbank. Today Donna will discuss emerging trends in data architecture. What's the next big thing? Just a couple of points to get us started. Due to a large number of people that attend these sessions, you will be muted during the webinar. And we very much encourage you to chat with us and with each other throughout the webinar. To do so, just click the chat icon, which is in the bottom middle of your screen to activate that feature. For questions, you will be quick to move via the Q&A section. Or if you like to tweet, we encourage you to share highlights and questions via Twitter using hashtag DA Strategies. As always, we will send a follow-up email within two business days containing links to the recording of this session and additional information requested throughout the webinar. Now let me introduce to you the speaker of the series, Donna Burbank. She is a recognized industry expert in information management with over 20 years experience helping 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. So with that, let me give the floor to Donna to get today's webinar started. Hello and welcome. Hello. Always a pleasure to join you. Looking forward to a great 2019. So on that note, if you have not joined us in previous series, welcome. And today is the beginning of a monthly series. We'll be doing each month on data architecture and various strategies and architecture. This is January in the beginning of the year. We thought it would be interesting to start with some thoughts on emerging trends in data architecture and what is that next big thing or maybe what's not the big thing anymore, which sort of passe, right? So we'll start with that, but please do look through the rest of the year and hopefully there's things of interest to you next month. We'll be talking about data strategy and you'll see we have some interesting case studies coming up after that. So please do join us on future webinars. Always a pleasure to virtually see you. And for those of you who do come up and say hi at some of the data diversity conferences and things in person, I always appreciate that as well because I see some familiar names on the call today. And it's always nice to put a face to a name. So welcome to those who come back regularly. So today, as I mentioned, we're talking about emerging trends in data architecture. And if you've heard me speak before, I'm always stressing the exciting time that it is and data because it really is such a link now with business. And I'm a business person at heart and I can link two of my favorite things, economics and data together. This is the time to do it because the world is changing. So although that's exciting and I'm a big old nerd and it's really fun to be in data management right now because technology is changing so fast and there's so much opportunity. But with that opportunity, it can be a bit overwhelming of where to begin. What are those things I should really pay attention to? What do I have in the corner of my eye? And what should I avoid that maybe other folks tripped on first as an early adopter? And maybe you don't want to make that same mistake. So we'll do that. But as always, I will try to, as Shannon mentioned, I run a consultancy as well. And we have customers in various industries all over the globe doing this. And hopefully I can provide you some practical value for what we've seen through all of our customers and what actually works and what we can actually do because I know there's a lot of theory. There's a lot of things we could do and read about, but what actually works. So we'll kind of stress that today as well. So as I mentioned, data is hot and data is hot because it's driving business. And this is an article a colleague pointed out to me a couple of months ago. And I love it because it really sums up so much of what I'm always stressing as well that we are in the data driven digital economy. And this is from the World Economic Forum. And they did an interesting study. So if you look even just five years ago and five years ago is not that long. It seems like just yesterday, the largest companies in the marketplace were selling stuff. They're selling things. So things like Walmart, you know, the classic thing seller, right? Or Exxon, you know, they're selling products and services. Now, if you look in 2018, just last year, I think this has changed even more so with some of the percentages. The leading companies are focused on data or technology or digital. And when I hear digital, I hear data because the digital is driven by data. And they had an interesting quote that now in this information economy, data is now literally more valuable than physical tangible objects. Even companies like Amazon that sell product, of course, really the differentiator is data and not only the amount of data they have and the amount of data they leverage, but how they use that data. The recommendation engine, you bought this, bought, you might like this. I mean, that's AI, that's big data. They're actually doing a lot of cutting-aims things with data, which really give them that strategic advantage, as well as things in the back end with supply chain, right? So whatever your industry, there is data associated, and that's a great time to be in data. And so I really like this quote. You can read the reference there if you want to read the full article. Some of the other things I'll be touching on as well, they actually go into a bit more detail on what some of the drivers are, even things like visualization. You know, it's not always the back end. Often it's the front end that's making it more accessible to business users. And so I kind of brought this back to my practice and what I'm seeing. And I just did a bit of a, you know, back to the envelope calculation of, you know, when we have customers from around the globe working with us, what are the main drivers they're seeing? I thought that might be helpful for you. Really across the board, 8 out of 10, or 80%, in some way, shape or form, we're doing a bit of digital transformation and typically use that word. And we'll go through and we'll talk a little bit more about the range of industries, but it could be a complete web-based application from a brick and mortar store. It could be, I'll talk a bit more, one of our company's customers is a small nonprofit, and they're making everything more automated and digital, because they don't have a lot of staff. So really, if you think that data management is not something you can leverage in your industry or that digital transformation sounds like a buzzword, and that's not for you. Think again, because that really is something that, yes, it's a buzzword. We love buzzwords in our industry, but it does not make it not real. I guess an analogy I like to use too. Old enough to remember the dot-com error, right? And everyone said, oh, that was a boom and bust, right? The dot-com error. So maybe stock market-wise, yes, it was a bit of a bust, but what isn't dot-com? Think of Amazon dot-com. It was a buzzword that actually became a living reality, and that is our life. I mean, brick and mortar stores are going away in replacement of dot-com. So digital transformation, I see really is that next generation of what we started at this dot-com. It's really making that literally more data driven. So I found that interesting and it's kind of fun to do these, because again, as I mentioned, if you are the type of person that loves data but also loves the bigger picture, I mean, I've said on previous calls, I was an economics major starting out when I was younger and being able to meld those two now in such a nice way that data is driving business, and you really can have a seat at the table really doing digital and business transformation. So it's a great time to be in the biz. This, I put together my own little visualization, mostly accurate. I didn't do the full Python model here to generate this visualization, but what I found interesting and why I love my job, I'll just come out and say it, because I get to work with so many different companies across the globe doing such different things. And what I also found interesting is the range of industries. So if you are in the biz as long as I've been, when we grew up in data management, basically it was finance, insurance, maybe government, but healthcare, right? Those were the ones that either had the pockets to do data management or had the need and regulation. So, you know, I'll be a bit of a cynic off in this opportunity, but often it was regulations that make you have data audit and things like that. So finance insurance, if you're in data management for over 10 years, you've probably been doing something in those markets. So the fact that that's the big old blue one in the middle is not a surprise. But what I am eagerly pleasantly surprised is some of the different ones you see there. So look at that big light blue on the lower bottom. And this is from our customer base. This is not an industry-wide survey, but it's probably a microcosm of what others are seeing. We're seeing a lot of nonprofits. I mean, one of our most successful customers is just won an award for the best managed nonprofit based on data governance and master data management in their Head Start program outside of Detroit. We have some museums we're working with. Folks, I never would have thought if you had asked me 20 years ago, would you be doing advanced data governance and master data management for a Head Start program? Probably not. And that's awesome. And part of why that can happen is, well, we can learn from the folks that did it before. I think most people can see the value in a lot of the tools and technologies. I'll talk more about that are so much more accessible to the average person. Literally, even sitting in their pajamas, if you are that nerdy and want to, you know, do some advanced visualization on the weekends with open data sets, you can. I mean, it's amazing. We're seeing education in universities where we talk so much about, you know, single view of the customer being customer driven. Well, they're customers, students, human beings. How do we actually determine better outcomes through data? Similarly with healthcare. So we've got several customers in healthcare trying to, you know, some of the typical, how can we get efficiencies by even getting a single view of our providers? You know, apply with regulation, but also once we get that down, how do we actually have better outcomes? How do we do better, you know, patient management? Manufacturing, I'm finding that very interesting because so much we think on the front end with customer. And so of course these manufacturing customers are also looking at customer because everybody has customers, right? So, but also when we think and think of Amazon again as an example, the back end supply chain. And this is where things like we'll talk a lot more about AI and artificial intelligence and machine learning are really transforming the business and just optimizing supply chain. One of our big customers in Europe is we're working with some analytics. And just because of their global scale, even a percentage point of a half a percentage point in efficiency, when you scale that can really make a difference. So we can talk about KPIs, we can talk about metrics, but this real, especially in things like manufacturing, a small change can make a big difference. So really monitoring those KPIs is completely data driven, right? We're always talking about data driven, but so many more companies are even that nonprofit that I'll mention again, because I'm so proud of they have KPIs. I mean, would you think that you would have your early childhood teacher with KPIs about your children when you think about it that makes total sense? How are our children doing on that test? Who's had their immunization? How are they doing at home, right? And having your dashboard, your teacher dashboard so you can really see that change. Media and entertainment, everybody. What I found about that one, I thought it was fun with my first client that PII or personally Identifiable Information was your avatar. I had never really thought about that. They did some online gaming and in that world, people know you by your avatar. So that was considered PII and had to be sort of masked and obfuscated in the database accordingly. So no matter what industry you're in, there isn't a way you can become more data driven. No matter what restaurant they're, they were sort of optimizing their restaurant and I mean, sorry, their menu and their supply chain to get the right products on that menu for the best price and having the right menu for the customer. So no matter what it is, you can use data in a certain way and use some of these tools will be talking about in a very innovative way to probably drive more profitability. Or if you're again, if you're a nonprofit or your school, better outcomes or healthcare, it isn't all about profit. It's about outcomes. It's really a great time to be in data and that's really a democratization of data management. I know this is an eye chart and I'm sorry, but what I had another slide and then I put this one back in because people, we had some lots of people like this kind of radar diagram. But what I found interesting is so what are people doing, they know they want to do digital transformation, they want to be more data driven. One of the things we start with in our practice is a bit of a maturity assessment. So where my analogy, I think it was an article on the diversity where they sort of quoted me with this analogy on the strategy. I'm a big runner, if you've met me, I'm really hyper so you're going to run my system. And you might come to me say I want to run a marathon. I'm like, that's fine, anyone can run a marathon. You can run a day and you already run a sub three hour marathon and you want to get into the Olympics, or are you sitting on the couch now and you just want to lose some weight and you really need to do marathon or you just want to jog a 5k. So that's sort of what we do with this idea of strategy where do you want to be because that that drives everything else if I really just want to lose some weight around a 5k. I don't have to train 20 miles every weekend if I want to get into the Olympics I probably do so that that's this blue line is where do you want to be. And then the rest drives how you get there. So I am seeing most of my customers with a fairly good sense of their business strategy knowing that they want to be data driven they're reading the articles they're seeing the success stories in the per that previous slide that people of all industries are becoming data driven, or I think people are have the question mark and hopefully, you know, webinars like this with diversity can help clarify some of that. So, what does that mean what do I do. Okay, I want to be driven I want to be digital got it what what does what do I do. And that's where that idea of a data strategy comes in. What is the data management do I need to do better. So I can become data driven. Yeah, next Amazon, right. So, one of the things that I'm pleased to see that often comes as result of that is data governance and we'll talk a lot more about this. You can often think of data governance as you must do this, or the regulators are going to come sue you or your customers are going to come sue you and you're going to get in trouble with the auditors. Yes. But it's much more than that now and it's really a lot more alignment between your business goals and technology and most of many of my customers that have data governance committees. And more it's the data innovation Council, or the data strategy team, or the data evolution group, because they really see that in order to manage data as an asset which is all about what governance is. It's really about innovation and evolution getting there together and collaboration. And I'm a big fan of saying if you get the right people in the room. You don't have to have as many rules we're all adults we want to do the right thing I don't see too many customers I'm hiding their data on purpose from each other. We're busy and new other things to do. So you get right the right people the room and often the rest sort of comes out of that. Master Data Management is one of those that you know what's all this new again and we'll talk about that as well. I want to be data driven I want to do the recommendation engines and chat box and understand customer segmentation and data analytics on that. So if I don't know who my customer is do I have one Donna Burbank or do I have seven people named Donna Burbank and they're all different people, or those seven versions of Donna Burbank. And she buys a lot of stuff. I'm probably not because I don't buy things. I'm just working all the time. Data Warehousing and we'll talk a lot about more that as well was that reports of my demise have been greatly exaggerated. Then warehousing is here to stay. Most of my customers have a data warehouse. Many are building brand new data warehouses. Yes, yes, that's still helpful for what they are. Are they done a little differently. Yes, are the tools that you can be more effective. Yes, can you build it in the cloud now and not have to have your own servers. Yes. So, not saying nothing has changed in data warehousing and it's the same as in 1997 started, but you can do a lot more with it. The use case is still there. We just do it differently. Similarly with the I huge reinserted or rebirth and that or continued growth and that I think the tools have become a lot slugger. I think the big change there is not whether people are doing it, but I think this idea of self service is really everyone wants to do it. Are you doing it but who in the organization is it just the bi team or is everyone across the organization. Often what is difficult to scale in self services idea of this metadata layer or semantic layers when a lot of people calling it now semantics means what does it mean what does the data mean. Yes, I see that there's a KPI there I see that there's a table there but what the heck is table 123 meet. So to make your data consumable, you really need that metadata layer so I need to work with me before you've seen some of these previous seminars. Often, you know, these aren't in a vacuum you can't do bi without metadata you can't do governance without metadata you can't do metadata without data integration right so they're all linked together. Data analytics is the hot one. And you'll see here I made this this is sort of again like the bubble charts fairly representative of the, the average of the customers we've worked with around the globe. In terms of where they want to be which is the blue line as well as where they are, which is the green line. So I think often the biggest there's a big gap in data analytics, not because people are not doing analytics but back to the marathon exam analogy is because where they want to be. They're running already but they want to get into the Olympics. Great example I'm working with a couple insurance companies now. They've been doing analytics since the beginning of the day of time, right. Insurance is analytics me to give extra worries and a lot of that. They want to actually be much more real time analytics I want to have automated pricing they want to be, you know, they really see that data is a differentiator. I want to really get to the next level of AI and automation and artificial intelligence. One of the massive insurance companies I work with is doing a merger between a big London based and a big US based insurance company, and the CEOs quoted in the big kickoff saying we, we merged because of our data assets the combined data assets of these two companies is the value of the company and the analytics we can do in the X URL analysis on that. That's proof that data is the new asset of this 21st and 3rd and so often when we want to do things like data analytics. We're often hampered by data quality. So, this is the case where it's not all bad I am seeing getting old been around for a while. I think it's actually better than it has been in the past I do see a lot of people seeing data quality as an asset they may have dashboards that they just like we have KPIs for data they have. Like our company you know how much sales do we have they have KPIs for quality how many empties do we have how many of these values are not within our business rules. So, but with that the more you monitor the more you see right so there is a gap from where we want to be if you're making business decisions on this day that has to be a lot better than it was in the past you can't really have mistakes. So that's a good and bad we're looking at it more carefully, but we're looking at it so that that's a great thing. I think we move along this little diagram data architecture is similar. It's a fairly mature field. I am pleased to see often when I go into an organization they do have a lot of physical data models and data dictionaries in place. Often what might be missing is sort of the big picture and again that sort of relates to governance and in the siloed applications again nobody woke up in the morning and said you know I'm going to make it so nobody can see the entire architecture of this organization. They're busy and didn't have that time to really step back and look at that but those can be so helpful even at a high level. And maybe the other gap we see, and maybe it was fine in the past when data was driven by it more so, but now that business people are more hands on with the data. This need for business level data models and glossaries and conceptual data models. Again what's always new again and seeing a lot of interest in that so again there's gaps there but gaps because there's more demand and not because people are getting worse I think people are getting better. With data asset planning and inventory. I remember back in the day when I was first starting my career and data management, being very scared to go into some of these big companies and I was doing a lot of metadata repository work back then. And a lot of our purview was could you just even let me know what systems we have, or what data sources we have we can't even get our hands on that. We would just spend months just doing you know automated scanning and trying to get that inventory together. I don't see that as much of course that exists. But I think people know what they have but they're wanting to optimize it. I think also now that we're looking at this more carefully and there's more regulatory as there should be, especially when it's customer data. More carefully at things like life cycle policies and retention policies, and then particularly with regulation do we need to store we might have it. What we not only for so so much data storage because that's so cheap now, but what is what is a risk where maybe we don't want to have the liability to have that data integration. And I think, you know, things like ETL often is very for a data extract transform and load for things like data warehouses are fairly solid generally fairly well documented which is a change I hadn't seen in the past people are definitely maturing there. But again, there's gaps because there's so much more opportunity things like real time data streaming. How do we integrate with the cloud these new technologies. And of course there's silos because again people are just, they have a day job and they have to get things done. So often that big picture integration is missing and when we look to do things like digital transformation. That's almost often one of the gaps we need to start looking holistically for us to organize metadata my favorite one. I think you'll see a big gap there because I think people know they need it, partly because it's a confusing word metadata sounds nerdy and techie. So I think people know they need it they don't notice like how I do see a fairly generally some fairly good data dictionaries and maybe some physical data models and things like that. And when it comes down to the business alignment with that and things like the leads. That's where people are there's some gaps exist but the good news is a lot of good solutions out there now. But again, like I said in the beginning, often with those solutions that could be confusing because there's so many options so I know that was a busy slide and we spend a long time on it but often that's one of the first questions I get when we go into a consulting and engagement is what are the folks doing so this is this is based on data we've seen based on our experiences so hopefully that was kind of a helpful to get to where you might sit as well. Um, this comes from a, what is it is a survey sorry need more coffee that we had done with the diversity was last year so we have sort of promised you the update this year for many many reasons. We had a bit of a gap in our traditional teams, but we will have a new one that will be out in March and we would love for you to give us your opinion on this so stay tuned. Data diversity will be publishing that shortly, and I'd be really interested to see how this has changed what we've seen in the past few we've done several over the years. Maybe the story doesn't change it's the mallet matter of volume. So maybe more people are moving towards big data in the past but the general stories have been saying the same thing so I'd be curious what that that is now. So the first news here at the top just makes me sad is that spreadsheets are sort of probably the one of the leading databases on the market. So it's a great I use them all the time. They shouldn't be at their place they shouldn't be your master data source for any organization they're great to kind of do analysis for, you know, one off type things but they are too often used for actual industry enterprise metadata data management. So one of the bottom does not bother me. Again, the stories of our demise are overly exaggerated relational databases despite the hype of they're going away they're going nowhere they're still the leader they're still great they may be moving to the cloud, but they're still here. What I found was interesting in this one there was more legacy platforms and big data in terms of what we're doing today, which means brain frame was a fine solution that worked right. Am I saying start something new with mainframe know, but they still do exist in there needs to be when we look into the future. Not a surprise here things I've been seeing as well. A lot of people are looking at these big data platforms in a way to really expand their horizons into more big data solutions I could do movement to the cloud so again, relational databases aren't going away. A lot of them are moving to the cloud as well. But the other thing I thought was interesting and not not surprising at all uncertainty is common because when you look at those lines. There are if you look at the previous ones a lot of clear peaks. There's databases and there's big data and there's a bunch of spreadsheets. But when you look at the future as people grow there is sort of an evening out, which I think is a very positive thing because one I can rant about data I don't, again, some of you are probably rolling your eyes to work with me don't get started. But one of my rants and I'll rant a bit more about it is some of the hype of, we have a new tool and when you hammer everything looks like a nail right and so all of us take many of these technologies are excellent choices for the right use case. And often I see customers saying well I heard we need real time streaming well for what use case or I hear I need big data platform as a classic one you might, or maybe you don't have big data do you have internet of things no everything you have is in a relational database and you're trying to say total sales by region in warehouse is fine for you and you don't really need a big data so you don't need to jump on every buzzword that's out there, evaluated yes because there's some amazing things. But so I think this this evening out is actually a positive thing. Okay, so in terms of height. I'm a fan of Gartner they often have some really great reports if you're not familiar with them. They're a historical analyst firm that has a lot not only data management where they're very popular but other products as well. And this is something they do in terms of where are we on the hype cycle, where are you in terms of peak inflated expectations and things that everyone things can do everything from save their marriage to make, you know, dessert talking at the end of the day. And then, when we realize it can't do everything we expected we have this trough of disillusionment person ends up working out the things that are helpful stay and you have this practical productivity for several reasons. A this is their proprietary methodology and be I sort of think this is a little fatalistic it just seems a little, you know, we have this hype and then we're in disillusionment it almost seems like. It's a little dramatic but I created my own, which is probably equally dramatic and my own goofy way, but it might resonate with you so I have my data management mountains of momentum okay that's bizarre but I had too much coffee one day, but I think it's a similar thing but with a bit of a different take right so there's certain things that the mark is hyping. I don't see a lot of people using it yet. There's certain things that are hype, and they're awesome, and they get me excited and we the things when you step back up what we can do and we couldn't do before are just amazing and they're going to help your company, get to that next level of see change, but they're pretty new. And that's why this mountain peak is sort of new. There's certain things that we couldn't do before and are awesome, but people just think they're even better than awesome, and they're just in the clouds. I'll give you an early high things like AI artificial intelligence machine learning is one of those. Are they helpful, yes. Are they everything that every vendor says they do probably not. So, that I'll kind of go through each one of these. The one ones that we overhyped and then we, you know, the curmudgeons in the world like me said guys that doesn't do that guys probably stuff that's up here right now. Guys that doesn't do that guys it doesn't do it doesn't do that. And go through much on the call for yeah we told you so. So, but it doesn't mean they weren't good ideas, we're now just using them in a more realistic way. So I live in Colorado I'm in a big mountain climbers. It makes sense to me that this is your kind of doing this steady climb up a nice hill and had you just kept walking you would have been fine you didn't have to jump the peak and fall back off. Just use the use cases for what they need to other things along that steady climb. I just reread a text it was talking about economic text I was assigned to be back in college or random walk down Wall Street, where basically tells you all the peaks and inflations of Wall Street, but just invest in the S&P 500. And just over time you'll kind of get your steady x% right. And that's what I feel about some of these technologies things we've been doing all along. And he's like architecture and data models and glossaries and metadata and yep they're a good idea and they will continue to be a good idea and how do we continue doing them. And many companies have as I mentioned in the maturity assessment, we're a lot better in the industry than we were, we might be over critical of ourselves, but we are improving and I think that's that steady climb of these like metadata that are kind of moving up. So that's how people go, you know, progressively. So that kind of getting better at what we do. And then on that note of of we've been doing things all along we could have spent steadily but there is technology I don't want to discount technology. Some of the things we've been doing all along. We can just do them a lot better now. So my analogy is a car, right. So we've had cars forever. Things but now we have electric cars. We have cars that are going to get a rental car. I'm always amazed that you know now they beep at you if you're going to hit somebody or you know that I would have that happen to me. But it's just again we've been doing things and I'll talk a bit more about things like metadata that we can just do so much better than we could. So hopefully, oh and that at the bottom things we really shouldn't have ever done and and maybe guess the convergence in the call that you shouldn't do that shouldn't do that people didn't we told you so you shouldn't have done. You learn from that or maybe we haven't but you know sometimes people do fall off the cliff and things we haven't. So we'll kind of walk through all these and hopefully it'll be a helpful analogy if not I had fun with it. Sorry. Okay, so this is and I look this is definitely one of the anyone who's given presentations probably goes through the same angst but I have too many slides I want to tell them everything. I'm kind of an excitable person so I do that so I can go through all of these. I'll go through a few things. Just the general statement that I've made before it is an exciting time to be in data management if you're not excited to do some more reading to just a lot of new stuff you can use. If you just sit back and think of. Okay so I live out near Boulder, Colorado and there's a lot of nerds here one of my nerdy friend I use nerd lovingly is a good thing. He works for NCAR the National Weather, you know, Atmospheric Association and his dad did too. So we all have the dad saying well when I was a kid and I had to walk uphill both ways they sort of had a data version of that. Where the dad said, did you know that the type of data you can do now with Python and cloud based solutions and open data says we literally had a building in Wyoming that had massive mainframes and we still had a percentage of what you could do he was so he's retired now and he's not able to do what he could he was so frustrated he said if I had the tools you have now to do this kind of analysis when you think of it you literally could sit in your pajamas and spin something up on AWS with some open data sets you downloaded from NASA and get some big data storage. You can you know there's real time data streaming and you can do self service analysis and analytics with things like Python and it's amazing what you can do so there's a lot of good resources if some of those words didn't make sense to you. And I've done it myself I mean I did a lot of this in university actually some of this is one of those it's always new again. But things have changed and things like well the diversity of course things like Coursera and a lot of the universities MIT and Harvard and all of the universities have open courses to take advantage of them. And it's there's a lot of great stuff out there so this and Canon will be a lot of the stuff we'll be talking about throughout the year. And one of these could be their own webinar. So open data is one I'm a fan of in fact I think I mentioned the intro. We have the Environment Agency of England coming on in a couple months as a case study, and they are doing data models and metadata and glossaries partly because they're publishing scientific data sets to the public. So, and again I got myself off track but to two folks in the organization that had actually met through open data because they found they were two people in the world that they, I forget what it was like weasel scat or something ridiculous but there was some animal that they both analyze the poo of. And they found that other person in the world that love that too so not only can you find people that are lined academically but you may find your life partner through open data I cannot promise that but again, there's so much opportunity but as with any organization when there's opportunities a lot of responsibility so that's where things like metadata come in. Um, the other one this over hype, and I will put AI and machine learning squarely in that category and this is a Matt will also off Twitter, I've seen this code come out quite a few times cracks me up. Half the time when companies say they need AI they really need to select clause with a group by your welcome, not to be an old curmudgeon. But again, so many, you know, I'm honored to be able to work with a lot of C level execs and a lot of big companies around the globe, and, you know, have a CEO come to me and say, what do we need to do for AI, right, generally, because I don't mind speaking truth to power. I said, this is you actually need that first. Right so let's talk about what you need to do an AI very well maybe and is it true AI or is it machine learning or is it we can have again the whole webinar and what is that. But this is a case I remember in my economics major, but I also have a computer science degree. And we did some tests with AI and machine learning and I remember just being fascinated by that. And had I more time in my life now I would probably just dedicate myself to that because I remember wanting to do more but we were limited by the tools in the data sets and what was available. But with things like the big data, and the fact that you can get these data sets to truly train. It really is truly a one of these seed change technology so. Yes, it's amazing and yes you should be looking at it, and yes you should take it with a grand assault that every vendor that says they do AI and every company that says they need to do and you may not need to or want to, or think of the maturity, you just need some better business intelligence reports to see how many sales you have by product and you may find out that I don't even have my product codes rationalized and I have different product codes in each country and I can't even do AI because I have crappy data right excuse my language. So yes look at it but it can be definitely over hyped. And one where hyped, I think has settled into reality, and this is where I feel old. I feel old, but then I don't feel old because I realized that time, they have dog years and we've technology years where each year is like 20 seconds I feel like, but I remember when I was at the break grand new thing, and everything and so somebody that still isn't everything can be solved with a data lake and crazy things that I said at the time and no one listened. Well, a lot of us listen folks like diversity folks but there was a lot of misunderstanding, you don't need data warehouse thing or everything can be put in the lake and you don't need data quality anymore because it's a lake and you don't need to, etc because we have lakes, and that just made no sense at the time and it makes no sense now. A lake is a great thing if you want to disparate data sources and you have video files and Internet of Things streaming data and you want a lot of raw data to be sourced in a staging where you can look at after the fact and it's induce some analytics and exploratory analytics real time you can have a lot cheaper storage. A lot of great reasons of a data lake but it is not the solution to everything. So I think people now the good news is, I'm seeing less of that. But what makes me feel old is suddenly, gosh, those those data lakes that's old school. That was like 10 minutes ago, because I think there's the, just like people said old data warehouses they never work. Well, this point of successful ones we also hear old data links, they never worked. And I think what frustrates me is you can put any single technology in the planet well X never worked well you could do anything badly. You can do things well and some technologies don't work they're not the right solution that we'll get to those but often it's how you do it not whether you do it or what it is so I'm going to spend most time on this if I have some time. So is I think a lot of the fundamentals are still there and I love this joke that it's kind of like the car there's now an electric car this looks similar but it's powered by Hadoop or reinventing the wheel. But a lot of things we've been doing for a long time and are still good ideas. We can just do a lot better with these new technologies. So things like machine learning and cloud based solutions and massive scale can now be applied to things like metadata, which we've been doing forever. And things like you could be again a little contrarian and a lot of other things that are bigger things like data monetization of monetization of data and you might say we've been doing that for a long time using data to make profit right so data strategy so you've never had a strategy before micro services not like you've never broken things up into small little areas and different technologies and use them that way right but but they are different now in this way we've improved. I'm going to walk through these because I find this interesting so this slide, I think sums up I feel kind of more the realistic view of where data lakes and master data and data warehouses kind of can fit together so our data lakes still valid and viable yes our master data still viable and where yes they just have their fit for purpose and so if you think of the data late as a great place to sort of stream data in this real time either through sandbox, or through lightly modeled non sandbox data that I'm really using for true analytics. And, but it's kind of disparate and large volume variety, all of that. That's fine. You also have data warehouses for some national reporting and things like that that have to be by definition cleanse. So both ways so master data yes it's hard to do is it valuable yes, because you need to do good analytics you need to throw data, and often master data can be fed into these modeling, you know, analytic environments. Should you have security and privacy whether it's on the lake or not. Yes, I mean one of my names will be protected for the innocent to protect the innocent. I remember one of my large international financial services clients we were talking about the lake and PII PCI and PI one of the junior developers and so I shouldn't be loading the credit card data into the cloud. Pretty will talk to you after the meeting and so just because it's in the club just because big data or missing the one client because question was, do I still need to worry about PII if it's in documents and not data and they said well as someone steals your credit card and they tell you well it was it was on a PDF. Did you care. Oh, it's still data right now where it is so and that's where that data governance box across the top can help is that you need the teams that are both your document management team and your analytics team and your warehouse team and your business users, all in the same room to really get that cohesive scope so that you can do the stuff in the pink or move or whatever that color is up top, you can do the true analytics and self service BI. When it's nice and clean and well organized with metadata you can do that. And that piece in the right the old fashioned you could say data models where what's the product and what's the account what's the customer still hold because as soon as a business person wants to do self service and they see a list of customers that don't own your product and you say oh no no no those are for marketing well that's not a customer and all those questions will come up again, even more so because more people So I don't think the good news is, I think more people are getting that I hear less of the hype, where everything's a lake or everything's a warehouse or everything I do see more than a maturity in the industry that people understand that the fit for purpose. And our I think some of the challenges how do you get their right ecosystem of these fitting together the integration not just the ecosystem of tools but the ecosystem of people, which is that data governance piece. Another thing again could be several webinars on ones on this where I do think is a true sea change is this idea of visualization. A to this really really fun, but it does also especially as more business people get involved. They want to see things in a more intuitive way and as you get these massive data sets, you can't have the report in the left that's total sales by reach I mean, and in just usability we're competing with things like cell phones and Netflix and we were just used to a very rich visual experience and people want that at work as well. So we see less of what's on the left which again was taken from micro strategy that's not a knock that was an older report from there was nothing wrong with that tool. But that's sort of what we were used to seeing when we thought of data. Now. Wow, the stuff on the right. And again, a lot of this can be downloaded with open data says we're doing graph databases we're doing you know heat map literal heat maps. And so that's another thing if you're interested in data and even have an artistic bent this whole competitions and visualizations and some of the vendors support that as well. There's a, I was lucky enough to hear him speak in London at one of the conferences data is information is beautiful. recommend that just there's this folks that have not in our traditional industry that are seeing this as well you know that are this whole art of how do we visualize these massive amounts of information on the planet is just a really fun thing to be in and really, and I would challenge you and I've done this myself I am so myself. I'm so used to sort of showing bar charts and radar diagrams and kind of a challenge for me when I'm working with the customers how can show this information in a different way. And there's so many resources that my my head explode that I've never thought of showing something in a way that might have been shown before our infographics I mean I've seen several of the bi tools actually integrate with the marketing infographics tools. So that's a great way to show information. metadata near and dear to my heart. So interestingly, we did some of these surveys we've had done in the past. Over 80%, I think that's actually low said that metadata is as important as not important in the past and part of the reason for that is more people are looking at the data and more people realize that this is a business asset. I think this is a what an area where the tools are really helping with technology so not that either of these doesn't exist in neither left or right or bad things. But I sort of group the one on the left is your more traditional metadata repositories the part that is involved are kind of a similar in both tools. So your traditional metadata probably has for your data dictionary of structures, your data lineage. Most of these with these more modern and they're called kind of data catalogs. I have that as well. I think some of it is a philosophy and tooling that on the left is I you may have heard me say this before the difference between encyclopedia and Wikipedia, where encyclopedia metadata is this is the definition. Are your curated data sets about shell use them we've had get a governance committees vet them, and these are the standards and guidelines and here they are on the right. And again, neither one of these is bad just think of the use case on the right is more if you are doing self service and you want a more collaborative approach, and you have trusted data sets but sometimes those trusted data sets are based on user ranking. This is the correct customer data center this is the open data I use that's really knocks it out of the park right. So that's a style, but some of these tools. And again, it may be an overstatement of what machine learning and I are, but some of the old school things we have to do I remember in the old days kind of mapping. X X X X X X X is social security number, or you know what an email field looks like and having to now with some of this machine learning and training, the systems can sort of figure that out and by the way this looks like an email to me, it could be called feel one looks like an email should we label it as an email and that's pretty neat that so kind of automating a lot of these kind of banal things and having more discoveries that you might have missed in the old way. I think also some of these glossaries they have more of a Google style search than maybe your traditional, you know, glossary type of so both are fine. I've seen customers go both ways and had trouble so some have used these many catalogs with huge success for analytics group. I had one customer that actually had a lot of problems with them because they wanted to really lock down GDPR data. And those the catalogs were at that time and the tool they chose with little to loose and it didn't let you lock things down enough. So think of that as you're looking at tools no tool is bad, but think of the use case and of course as with all tool they're kind of merging so, but just give that some thought. Data governance similarly is where I'm seeing a huge evolution, less the top down you must do this. Clearly that's part of what governance is, but more repair it unless with the stick, because once you focus on business innovation and data strategy and we're transforming the business digitally. I tell my customers this and they often don't believe me, you have people scrambling wanting to be on this committee, because you're seeing people see these are the system makers these are the people looking at data for not only how we manage it for new opportunities. And I think part of that is this increased interest from business users, not that data governance ever should have only been an IT thing, but the past often it was. The change is that business people are driving governance and they're more involved. And it's a collaboration between business and IT which I think is a great, a great advancement. And so in terms of traditional approaches that are gaining buy in and we're back to kind of our momentum. We've tried and true fundamentals, or I think keeps surprising me are seeing a resurgence. I've had so many customers come to me and want a data model that surprised me or not surprised anymore because it happened so much. Or architects or even governance is in that category master data management, we've been doing this for a long time, just like the wheel on a car wheels in the someday we may have hovercraft and not near made wheels. But the, you might have better wheels now, but you still need wheels, and these are kind of the foundational things that don't go away because they're really getting at your core definitions. I know a busy diagram but I wanted to do that. So, I think partly because more business users are looking at data, they expect to have some of this semantics and definitions so really fun for me because as you know I'm a fan of data models. I have had folks from scientists you'll see that the environment agency when they come on to early childhood teachers were up on a whiteboard I give them like three slides of what a data model is. And I can do not they're doing boxing lines and the relationship diagrams with cardinality and all of that it's an easy thing to understand like kind of teacher being more than one classroom and wouldn't that be and they'll even correct me shouldn't be a non knowledgeable feel I mean they get it right so I work with a water company the engineers got it marketing. I mean, people will understand a data model at a high level. Things like data flow diagram is one of my best favorite quotes I was working with a chief marketing officer right just think of that a chief she's very high level. Marketing is not general you think is wanting to get deep into technical designs, she said, I never thought I'd use the word data flow diagram in my life, but that's exactly what I needed to see why my campaign data was wrong because she knew her data wasn't great. But we had she didn't build the data flow diagram, but we showed it to her and it was a way she finally understood why the data wasn't flowing right. Well, fast and things like data lineas and cred matrices where data is created read updated deleted where I used one of my favorite quote we're trying to explain why we needed this finance, finance of all people. I mean, are you serious you're not doing this already. We couldn't get away with that finance I don't know how much money we have it somewhere, not sure the lineage of that. You go to jail for that right. So, also some of these traditional enterprise architecture diagrams when you're speaking to the business again don't take a year to do them you can do some of these in a few days or in a workshop or an afternoon. What are the business process models where is the data use in the process what are the business capabilities using this, especially that last bullet. So many companies now are doing data as part of transformational digital change which involves culture involves a change in how you're thinking, so you cannot do that in a vacuum, and you need to link this to what business capabilities do I have today what business process do I adapt today. And I don't want to just take my brick and mortar processes and put them on the web. That's not digital transfer me. So especially when we're doing that now. You want to lend your data to these bigger picture items and I've been using them successfully and people, especially business people love them it's often it that pushes back really I got to make it so simple. Yes, you want to communicate with other people. That's the goal of these. MDM is a big one that's coming back as well, especially everyone's trying to be customer centric right so how do we get that single view of the customer and not only know that stuff across is the one step on cross and not 16 of them. But how many persons he had and where he lives and what his occupation is and he's a loyal member and he finished the anger theme scheme marathon, which is one of my favorite races. And that's hard to do, partly because you need all of those things we've been talking about the data models and what processes touch this customer etc etc, but more and more people one of the first things I notice with my customers. I wanted to digital transformation I need to start with MDM because I don't have a good sense of what my I want to put all of my products online and have a customer loyalty program. I don't have a great view of my product list, and I don't know my customers are can't do either of those without good MDM, which they'll be product MDM customer. A couple where we're kind of the hype up the climbing up the mountain hype proceeds implementation I would put blockchain in that category seems interesting hearing a lot about it I on and correct me in the comments if you think I'm wrong. I don't see a lot of people actually using it for business, other some of the things we've been using. Great thing to read up I'm still kind of it seems interesting kind of that core technology, kind of the distributed ledger, again, a whole webinar just on that. Interestingly, and it could be, you know, maybe the diversity people aren't doing this and other areas are I just have not seen a lot of it. At that time only seven point will almost eight percent. We're kind of looking into it actively or I think even smaller percentage we're actually using it. Maybe next year when we do this it'll be a different number we'll see so take that survey when we have it out here your response. Some things that and again I am getting close to time and you're lucky because I could probably rant on all of these. I love to complain right so some things we might not have ever have done and in you might disagree but yes there are the agile folks and we don't want to do any documentation ever. That was never a good idea. And then there's been too many teams that sort of want to skip the documentation phase and skip data models skip glossaries and things like that. Don't skip them just do them in a better way data lakes we talked about that study replacement for a data warehouse is augmenting. Well we still do this third when just buy a tool and it'll fix everything know it won't you need to use the tool right correctly. And this is not going to do with data but open floor plans who came up with that I'm trying to do analytics my 16 people right next to my face. Not a fan of that one so I think I've heard other feedback on that and I'm not alone there so. So in summary, what is the next big thing what if Donna Burbank had a crystal ball what would things look like I don't know. Here's some of my thoughts I think the rise in digital business and digital transformation will only increase. But that's just a trend. That's a great trend we want to see more of that self service I see as another trend we're just so used to having technology our fingers fingertips and in being able to be more active in it. It's hot so everybody wants to have a finger in the data game. And so to be more of that, hardly as a result of both of one and two of the first two bullets metadata management will continue to grow and I think there will be more automation more self service. And I think there's a place for some better visualization. A lot of this visualization we do with analytics can we put on metadata we're better. I'm still not a fan of a lot of the data lineage I see. I think we just it's a hard thing to visualize but we're getting a lot better visualization so visualization is for I. I mean we can do so much was just virtual reality and you know visualization again we're visual creatures I think that's even going to get better than we see. And I sort of key AI machine learning because it's overhyped. I don't see that going away partly because there is such massive volumes and real time accessibility of data, which is a great fuel into AI and machine learning. But I don't think these foundational technologies that I mentioned like data models and life cycle management and business process change and all of that. They're not going away. You need that as part of that foundation to really treat data as an asset. I'm interested in your thoughts and predictive predictions in the, the comment section and just quickly before we go. Next month we'll be talking about data strategy, please join us. And we do this for living if you want to help us know. So that we want to open it up for questions and comments and I'll pass it over to Shannon. Thank you guys for kicking off the series and you're with a great topic in presentation. Just a reminder, I was going to follow up email for this webinar by end of day Monday with links to the slides and links to the recording to all registrants. So don't a nice spider diagram on slide eight. Do you share that for people to take their own maturity assessment. We generally that's part of our kind of gore best core best practices we walked through it with you. There's about 250 question 200 questions and we so no we don't just provide that as a download unfortunately, I will recommend though there is sort of the CMI has sort of a one this a little higher than the one we do that you can sort of purchase things like 100 bucks and then some of the other, even just going through things like the data management body of knowledge or Damodian box. I'm kind of looking through that and saying what am I doing and what am I not so even without a detailed questionnaire like we do. I think you can kind of do some of your own with some of those two right now. So do you really think the trend for cloud will continue to expand it just is it is just a server that someone else owns the cost cases that I've seen are not that compelling long term. I think it will. Yes, because it may not be sexy like it was but if I use kind of gardeners plateau of productivity. I think people are Stephen myself I just expect things to be backed up to the cloud my computer crash the other day and a lot of my files around the cloud and it was great. Did I really think about that. No, and that's almost because it's so ingrained think a lot of the smaller companies to who don't have their own servers and don't want their own servers can really leverage the cloud for scalability and testing so I would agree with you it's probably not in the new hot new technology stage but I don't think it's going anywhere I think it will only grow. But I don't think your own server goes away either. I think a lot of people want that especially for security and like that because it is somebody else's computer. So there is a question here you know an inquiry can you share a larger version of the back capability model with data overlay on slide 22. Um, I can do that also. I think last year, we did an enterprise architecture. An enterprise architecture webinar and we went into detail on that so on diversity everything's on demand. And we can if you, I think your email will be here and we can I can send you on individually. But you might be interested in that webinar because we kind of went to a lot of people how you build it. We have a bigger picture there. Forget what month it is but it's out there somewhere. Well, and we that brings us right to the top of the hour but let me see if I can slip in one quick question here. What are new innovative methods for creating data model from modern data warehouses. Not a quick answer I don't think but maybe we're going to test you heard on us. Um, well, just to a quick answer I think from the bottom up there's a lot of you know if you want an existing data warehouse. There's a lot of reverse engineering tools that aren't really new but lots of people don't know about them. I think just quickly on the top of new ways I think a lot of what is designed thinking and kind of the young hipsters. And I'm able to do a lot kind of whiteboarding and in that technology, especially if we started a high level in workshop things you know a lot of we love to tease millennials but they kind of are good at some of these clever ways to innovate. And so I've been able to just quickly and some of these design thinking workshops come up with at least a rough draft warehouse model, a lot faster than our six month, you know, waterfall that we used to do in the past. So might be a good way to kind of think outside the box and look at some of those techniques sometimes. I love it was on a thank you so much as always for this fantastic presentation. And Donald the enterprise data world in March. I'm very excited if you want to meet our person and you're hanging out in our new community, which is also community. And it works at that place for data people. I love it. Great. I'm excited about that. So anyway, well, thank you again and hope and thanks all of our attendees for being so engaged in everything we do. We just love it. Thanks for all the great questions and the engagement. And again, we hope to see you February 18th in the next month's webinar and hope everyone has a great day. Again, Donna, thank you. Thank you. Thank you.