 Hello and welcome my name is Yannick Kemp and I'm the Chief Digital Manager for Data Diversity. We want to thank you for joining in the latest monthly webinar series, Data Architecture Strategies with Donna Burbank. Today Donna will discuss building a data strategy, practical steps for aligning it with business goals sponsored today by CLICS. Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. For questions, we will be collecting them by the Q&A section or if you would like to tweet. We encourage you to share highlights or questions by Twitter using hashtag DA strategies. 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 on the bottom right hand corner of your screen to activate that feature. And as always we will send a follow-up email within two business days containing links to the slides and recording in this session and any additional information requested throughout the webinar. Now let me turn it over to Dan CLICS for a word from our sponsor. Dan, hello and welcome. Hello, Shannon. Thank you very much. Appreciate the opportunity to be with you today. I trust you can see my screen, okay? Looks great. Wonderful. Thanks. Hello, everyone. I'm with CLIC. I'm with the data integration division at CLICS. And what I wanted to do is just precursor Donna's presentation and give you a little perspective as we see it in terms of building a cohesive data strategy. And first, you know, if you step back and look at the landscape over the last decade, clearly there's a tremendous move to cloud. And so the requirements for your data strategy are changing. And some of the key drivers here, certainly the movement to cloud, but specifically, you know, organizations are looking to leverage the cloud for a legacy app modernization, moving data there, not so much a lift and shift perspective, but replicating that data from core transactional systems that may still reside on-prem. So they need to get that data there fast with minimal latency so they can build new applications and benefit from the elasticity and scale of the cloud. We've certainly seen a complete rise in data warehousing in the cloud. Organizations are looking to new cloud-borne technologies like Synapse and Snowflake and BigQuery and others. They're looking to reduce the cost. They're looking to reduce the complexity. But they're also looking for, you know, better ways in which they can move away from the traditional enterprise data warehouses of old to really better meet business requirements, be much more responsive. And part of that comes from automation. And I'll talk about a perspective there. The final requirement we're seeing is what we call next-generation analytics. How do I analyze a broader set of data? How do I apply AI machine learning? How do I pull in sensor data, structured unstructured data, do more real-time analytics? Again, this is where data leaks come in. This is where technologies, streaming technologies, like Kafka, come to play. So all of these are kind of key requirements that we see. And it's really kind of changing the data architectures. We're moving away from, and very rapidly moving away from that traditional, you know, on-prem data warehouse moving data into a warehouse in a very rigid, structured way, evolving to more of a data lake or data warehouse strategy, or both. And more and more, it's both. What may have started with Hadoop on-prem and extending to a cloud warehouse where the data lake is feeding the warehouse, moving more toward really blurring the lines and more of it's in the cloud than ever. Both data lakes, data warehouses, this notion of a data lake house where the data can reside in the data lake doesn't need to be structured. It can be structured on the fly for analytics. And finally, kind of this notion of moving to more real-time. How do I leverage technologies like Kafka to do distribution to a wide variety of different use cases, real-time visualization and analytics like broad detection, feeding data lakes, feeding data warehouses, feeding operational data stores. So the architectures are really evolving, really to meet the rapidly growing business challenges for fresh data, faster insights, faster decision-making. You know, from our perspective, one of the key things here is how do you keep up and meet your evolving data architecture with the data integration strategy that helps get you there. So you really need to think about how do you modernize and importantly automate the process of data integration and building these pipelines. From our perspective, it's really about streaming data pipelines. How do you take those source systems, generate changed data streams, which provide that fresh data to consumers, change data capture as a mechanism to generate these data streams, send them over the network to your cloud, to your data lake. It's a very efficient way in real-time to do that. But also how do you build analytics-ready data for a wide variety of different use cases. So, you know, the first one is how do I create an operational data store or how do I move it into a real-time database that is completely in sync with my transactional system or how do I publish these changed data streams into Kafka for a variety of uses. So this is about taking committed changes on those backend systems and replicating those in real-time. Another use case is around data warehouse and specifically data warehouse automation. Now, it's not good enough to just think about moving to a snowflake without changing the way in which your data warehouse is being developed and maintained, right? Taking that traditional ETL approach of over the last 20 years and just replicating that with your new snowflake data warehouse, you're not going to get any of the agility or benefits that you hope to. So you really need to think about, how do I take a model-driven approach to defining my data warehouse? From that, the ETL code is generated. It's pushed down into the engines like the snowflakes or BigQuery or Synapse for execution and continuous updating. Similar approach to data lakes, right? It's easy to throw data into a data lake. It's much more difficult to have a conformed data set that's analytics ready and easy for a business user to find and discover. So being able to do the merge capabilities on all this data and importantly being able to catalog the information. So providing a catalog from a data engineering perspective, as I generate these data pipelines, I'm automatically registering those data sets into the catalog. I'm automatically applying some governance like detection of PII data and remediation or masking quality steps as I move that data to ensure that the data is going to be fit for purpose, making it easy for business users to shop for that data, to prepare it if necessary to a derivative data set. And most importantly, not just finding the data, but how do I consume that data? How do I provision that data out for consumption through BI tools like CLIC or Tableau or Power BI? AI or ML tools, data science tools, whatever requirement is needed. And being able to go from the catalog to any BI tool like a Power BI or CLIC or Tableau, you need to be able to do that to meet the diverse requirements just on the BI guide. So this is our perspective from a data integration landscape. The key things to consider, one is real time, moving to more change, capture and movement. It meets the business requirements of fresh data. It's a very efficient way to move data into the cloud, but it needs to be together with automation, right? It's not good enough just to move that data into the cloud. You need to be able to provide continuous updates to be able to resolve changes to source systems in an automated way, push down processing. You need a solution that serves a wide variety of different sources and targets, and those targets seem to be continually changing. You need to provide a catalog that will also govern, and you need to be able to provision this data out for whatever use case is required. And what I've been talking about here is kind of that, from the raw data to analytics ready data. I'm not really talking about the other piece of CLIC, which is about analytics. CLIC takes the analytics ready data in the catalog and goes the full mile using augmented analytics, natural language and conversational analytics to surface insights immediately from that data set. So, you know, our customers go from complex systems like SAP or mainframe data on on-prem, move that data into the cloud, may move it into a warehouse or data lake, bring it directly into CLIC analytics and our surfacing insights, doing that all from a point and click perspective. This is where modern organizations need to get to to start driving better insights faster. The final thing I'll say, you know, I don't envy enterprise architects. The only constant is change, right? We spend a lot of time with enterprise architect teams and it's usually the people in the organization that start along this journey. Their whiteboard looks a lot like this. There's a lot of moving parts. There's a lot of change. So building out a data strategy and an integration strategy that's designed to support you today and in the future, because what you're building for today will definitely change next year. You know, look at the past 10 years and the rapid amount of change and different technologies and approaches. You know, helping to build a future-proof architecture and planning for change is absolutely essential. So that's our perspective with that. Donna, I will stop sharing. I'll hand it to you. Dan, thank you so much for this great presentation. And if you have any questions for Dan or about Click, you may submit them in the bottom right-hand corner of your screen and he will be joining Donna in the Q&A at the end of the presentation. Now let me introduce the speaker of the Monthly Series, Donna Burbank. Donna is a recognized industry expert in information management with over 20 years of experience helping organizations enrich their business opportunities through data and information. She's currently the managing director of Global Data Strategy Limited where she assists organizations around the globe in driving value from their data. So with that, I will get out of the way and let Donna start her presentation. Donna, hello and welcome. Oh, you're muted. Donna, you there? You're muted. Can you all hear me? If I'm not sure, I'm heard. Yep, I can hear you. Okay. All right, Donna, I see you're still muted. I'm not sure. I'm going to risk it here and Donna, I'm going to unmute you. Are you there? Thank you. I was hoping you would do that if you could move the slides as well. For some reason, the interface is frozen. Okay. Thank you so much. So as Shannon said, if we can move to the next slide, apologies for that technical difficulty. And I no longer can see the slides. So I'm going to trust you along as we say, move slides is always something. So this is the first, the second of a Monthly Series where each month we do have a different topic. Today is the topic of the hour which is data strategy. And glad so many of you, I see some familiar faces are able to join. As always, all of the previous, if you missed last month's and you wanted to catch it, you can see everything on the Data Diversity website. And there is a full agenda coming up for the rest of the year. So we hope you can join them as well. So if we go to the next slide, just a little bit on what we're going to cover today. So data strategy is one of those things that can seem like a daunting task, just the word strategy. It feels like it's just, you know, overwhelming or sort of a high in the clouds type of thing. But so many companies, and we'll talk more about that because data is hot. More and more companies are looking to develop a data strategy. So what this webinar will try to do, and if you've been to my other webinars, I tried to do this is really demystify some of these topics. As Dan said, we don't envy the job of a data architect, although I kind of do because it's fun, but there's a lot to cover. And so what I try to do in these webinars is keep it practical, keep it concrete, and hopefully share some experiences of what we see in our practice to really keep it simple ways to get started on what can be a daunting task. So if we move to the next slide, unless you've been living under a rock, you've probably heard about the data-driven business. And that's a great thing. It's a great time to be in data, especially for folks like myself who love to see themselves as both a data person but also a bit of a business person. I mean, my first degree was in economics. I really like to be able to blend the two. And this is the time to do that. I apologize for the data quality people on the call. There is a typo on this, but the data scientist being the sexiest job of the 21st century. I think that should be data architect, but I'll follow up with Harvard after the call. But you've probably heard that quote, right? Data scientist, sexiest job of the 21st century, because data is hot, because it has business value. Not fit in for the curmudgeons on the call. I think a lot of us could say it's been driving business for decades, right? But I do think with some of the tools and Dan captured it in the beginning, there's just a lot more opportunity and some really fun tactics that you can use to really drive that business. So going to the next slide. One of the other interesting pieces of it, and click is definitely the sponsor is part of this as well, is the idea of the citizen data scientist or the self-service data analyst. Because as I mentioned even with myself, I sort of went from data person blurring into business person. A lot of people are going the other way, right? So being a business person and saying, hey, I want to get my hands on this data, especially with some of the tools that are more slick and easier to use. Having these sort of unicorn people that can wear both hats and be both a technical person and a business person, if you are one of those unicorn people, purple people have heard a lot of different words to describe that type of person, this is the time for you, because there's a lot of opportunity in data to really help drive the business, and that's where a data strategy really sings. So if we move to the next slide, on that note, if you are this type of person and you want to have that quote seat at the table, this is the time to do it. Because there are a lot of opportunities through data, and I know in our practice, C-level people, business people are dying for someone who can really explain to this in a concrete way. So gone are the days where, and I've been in those days where you're locked in the server room in the basement, although maybe some of you still are and calling in today from the basement, we can be in the corner office with the windows, because people are looking for a way to really demystify data and put it in practical terms, and that's what we try to do in our engagements, but I think that's something you can do too. Instead of only looking at the tech, also think of the tech applied to business benefits. So moving to the next slide. What is a data strategy? I mentioned, firstly, I am a data architect by heart, and we love to start with our definitions. We could put this in the business glossary, right, or their data model, but terms are important. The words we use have a lot of meaning, and as I mentioned, strategies are becoming more and more the norm, and I've had sort of people ask me, well, isn't that just data management renamed? Haven't we been doing this for a long time? Isn't the data strategy just sort of understanding what we need to do with our data? And there's some truth to that, and even what we call a strategy, I've seen different, different, different from company to company. So, but what I see, if you go back to sort of the Miriam Webster dictionary definition, I think it sums it up well. When you think, look at the words that are used when we talk about management. Judicious means we're kind of managing, conducting, doing, organizing, and I think a lot of us in data management are good at that. We're sort of putting the ducks in a row and organizing the books in the bookshelves. That doesn't instill visions of grandeur, right, or really business strategy. So when you look at strategy, it's plans towards a goal. It's achieving evolutionary success. It's the art and science of meeting the enemy under advantageous conditions. That's a little more big picture and more visionary. And it is more of that business lens. And so I think a lot of us on this call are typically architect types to put on less of our, this is what we need to do. And these are the tasks I need to accomplish today, but put on more of that truly strategic hat and think what are the goals and objectives and vision that we want to align with this company. And it's a slight difference, but it's a big one because I think sometimes we are students that can imagine, oh, that's not going to work. And maybe it won't, but I think of why and what we could do instead to really meet that goal. So if we move to the next slide, data, as I mentioned, is hot. And this is though Shannon generally sends out an email after this and there's a link in the back for a white paper. We've been doing this yearly. So this quotes from both the 2019 and the 2020 paper that consistently say the same thing. The majority of organizations do see data as a strategic asset, 70%. The set-up is probably even higher. That's from the 2019 report. And 68% are looking to save costs and increase efficiency. So you might say there, I think that's always true. I'm surprised it's not 100% who doesn't want to save money and be more efficient, right? But that's sort of the traditional way and we'll talk a lot about that in this call of how you can really show the value of data management. A great data management protocol is very cost efficient and very lean. What's interesting to see and probably no surprise in COVID times is this rise of digital transformation. And one could say that's a buzzword. What does that really mean? I think we're all seeing what that means. We're all online, right? And more and more companies are very quickly moving online. It was interesting in our practice. Obviously, when COVID hit about a year ago now, a lot of us were sort of wondering what was happening. I was really impressed to see that a lot of the companies we had done data strategies with the year before really turned on a dime and were able to become digital very quickly because their data was in great shape. We've got some great success stories around that because a data strategy is the foundation of a digital strategy. I really see them intertwined. You can't go digital if you don't have good online view of your customers, your products, your deals, et cetera. And so, this has always been hot but you'll see in the survey sort of has increased 11% over the previous year. I am fairly certain if when we get the 2021 report, it'll be even higher. So, it's sort of interesting and probably not a surprise to any of us. So, if we move to the next slide. Oh, I think we... No, that's fine. So, there's different ways you can really look at a business strategy and a data strategy and twined and both are valid but I do see sort of a change in the industry. You can sort of see on the left of how do you optimize your business through data? How do I become that elusive data-driven company? And again, you can argue companies have been doing that for a long time. The companies that really understand their data are no surprise generally some of the top ones on the Fortune 500 list, right? So, how can we be more efficient? How can we eliminate manual effort and improve efficiency as important to think about how do we grow revenue? How can we have better marketing campaigns by understanding the 360 view of the customer, et cetera? And so, doing what we do better. What's interesting is I'm having more and more companies come to us say, I not only want to be a data-driven company, I want to be a data company where data is the product and we're really trying to monetize data where there's entirely new business models where you might have a spin-off of your current company just focusing on the data. They go to Facebook where, yes, they're a social media company, but really they're using the data for advertising, no social commentary there, or an Uber, right? So, yes, they're a taxi company, but really they're a data-driven company. They're using data from locations. They're using airline data to see when airplanes land and how many cars should go to the airport. I mean, they really are a data company and data is the product less so than maybe the product that's being sold. And so more and more companies are really looking to that and it's something that when you're doing your data strategy, look at the data itself. And neither one of these is inherently better than the other. There's a lot of profitable companies in both categories, but I do think it's an interesting trend in the industry that we kind of have both of those options now. So, I seem to be able to move my own slides now, Shannon, so that's great news. So, as we look at this, again, I touched on that before. So, this kind of interdependency between a business strategy that's going to drive your data strategy, do we want to be a taxi company and want to be an Uber, or are we and what's exciting being in the data industry, do we have data and we need to look at the data and say what sort of business could we generate from this? What do we have information that nobody else in the industry does? We had an interesting use case from, it was a manufacturer in Latin America and they had big trucks that were sort of delivering their product across very rural areas and what they sort of built was a handheld app that's sort of like a ways or a Google Maps or whatever but it was particularly for big trucks going over rural areas which have a very unique use case. There's weight restrictions, a lot of these rural companies, roads can't handle it. So, they actually created a product of data they had that's very different from their current business, right? So, that was a way they looked at their data and said what new things can we do? It was an interesting use case. So, if we look at this is a framework we have used and if you've been on my webinar before this may be familiar to you but it continues to be a really helpful way that we use for our engagements to really think of how we ticked all the boxes. So, where we always start and what I just spent a bit of time on is that business strategy. What are we trying to do as an organization and how do we align that with data to really make that thing? So, that's sort of the top down and then we do the bottom up for what data do we have and again, it could be that what data do we have to have a new business model or we have a certain business vision is our data in a state that can help us achieve that vision and we often start with some sort of maturity assessment to see. We don't have any data so that's going to be difficult, right? So, how do we really think that through? And then we sort of move up and down the stack. So, if you look at sort of in this, you know some of the bottom areas if you're familiar with the data management body of knowledge should look familiar to you. They're just sort of loosely aligned with that of how do we integrate that data and Dan talked about that in the beginning. How do we get the metadata around that data so we know what it means, where it is how it's sourced and then the next line sort of master data and warehouse. How do we start to leverage that for a strategic advantage, right? So, how do we make sure that there's master data so we have that single view of product, customer. How do we make sure that's of high quality and then what do we need to use BI or advanced analytics and machine learning, etc. And the reason we look at all of these is because you really do need to touch on all of these but not all at once. And that's when you start to build a strategy roadmap of it's that prioritization. The one other layer that's key to all of this is that data governance later. And you'll notice we call it data governance in collaboration because I'm a techie person. I'm proud of it. I've been in the industry longer than I want to admit. And even though I love the tech, the more and more projects that are successful that I'm part of is the people part. Is there a culture of data management? Are business people asking these questions about their business, about their data and wanting to drive analytics? Are people understanding that the data they put in drives the data quality there in analytics? And that's the hard part. It's really a cultural change. And the more people are bought in that they can see that the data they're using or managing or producing analytics on is driving either the macro business strategy or the micro business of what they do every day. It's going to make their job easier. That's really where governance becomes an obvious, of course we need to do this rather than trying to push people along and enforcing people to do things. And that's often, unfortunately, a negative connotation of governance. Oh, people are going to comment on policies and tell me what to do. Well, that is an aspect of it. You need to finance those rules around this as well. We sort of expect that. But the more you sort of understand the why and people can get behind it. So what's key to a strategy is to get all of this, the people, the process, the tech, and then the why of the business and getting that right priority. And why we always show all of these is because the answer is different every time, but a company might come and say we really want to do next generation machine learning and analytics. We might find, well, we need to get there, but the quality right now isn't in place to get there. Or we want to do more marketing campaigns, but we don't have master data around our customer. And we don't can't get there before we have the right architecture in place, etc, etc, etc. So they're all tied together. And the key to a strategy is really what weight do you put on each one and the order and the priority. And we'll talk more about that. So the way we approach our strategy and I'm giving away our cold secrets though, but feel free to use them. That's why we have these is good to take a structured approach. And this is our methodology when we go into clients. So there's probably no major earth shattering strategy. Most consultancies have something like this where starting with a very clear approach, whatever business goals and strategies and write them down. It doesn't have to take hours, right? It could be a white boarding session or a virtual white boarding session in today's day and age of what are our goals? How does data help with that? What are external companies doing? Just kind of, we've got some tools that kind of can document those really quickly. And then what data do we need and what data activities do we need to do to get there, right? We want to have better marketing. We need a better data customer data master and a better data analytics around customer might be kind of an obvious one, right? So knowing where you want to go and then assessing current state. If you have access to a data management maturity assessment or can bring in a firm that can do kind of a third party one, great way to know where I am today and where I want to head. You know, I want to get in better shape and do I want to run a marathon and start training now? Or do I just want to lose some weight and walk around the block? And then am I already an athlete? Or am I starting from scratch, right? So, you know, kind of the basics of what's that gap and not everything in every area. Maybe again, based on those business priorities where to start. You get a good sense of your technology landscape. Often these are missing when we come in and again, start. I'm a big fan. One of my, I stole from one of my colleagues that basically she said, women don't zoom out, right? Which I interpreted as having that conceptual layer of an architecture. Of course, that's what everyone would think, right? But that high level view of can we even, again, just don't be afraid to start with a white board rather because this could be daunting to really get a full inventory of every single system in your landscape. But actually getting some of the key architects together, key people who are doing these systems and do even a white board session how these things fit together and quickly get to the so what. Yes, having a metadata repository with all the lineage and everything is valid, but it's really helpful to start with these high level pictures and then the why and I, you'll see it's kind of small to see, but we often even put if there's a manual process, put a little picture of a person if there's three analytics tools and you want one show that there's three tools. It's just kind of a nice way to zoom out. So that's sort of the current state and then we go into future state and just obviously the future state architecture. How can we make that more streamlined? How do we consolidate? How do we put in tools like master data and metadata and lineage and all of that, but and don't forget the people side. So what is that right governance and organizational structure around data that you need to put in place? I saw a comment there, culture eats strategy for breakfast. Both are needed, right? You can have a great strategy, but it's not everybody's bought in and people don't and or don't know how they fit in with it. That's often we see will have a lot of people say great, I know we want to be data driven. That sounds awesome. What does it mean? What do I do? So that's where that actionable roadmap comes in. What do you do? All across the organization. What does the business do? What does tech do? What does management do to help drive this organizational change and really actionable quick things that people can start to see benefit. The other thing that can be problematic with the strategy because it seems visionary and big picture is maybe people feel too far out. There's a tactical aspect of data strategy rate. What can we do in three months to start to show the value and see if we're on track, right? You might want to redirect a little bit in three months and then communicate, communicate, communicate and evangelize across the organization. I did a brief in marketing in my career and it was probably, again, a techie person but that was probably the most valuable part of my career. You learn so much about do it once and communicate six times, right? People are busy. You might think you've said it, they might, they probably won't pay attention until they've heard it in different ways and it's applied to them and so often in tech which is on to the next thing, next sprint, the next deliverable. Make sure whether it's either you or a related team, again, work with your internal marketing, work with change management to make sure that what you're doing is communicated and everyone understands not only the why, but the what in a very actionable roadmap and how their roles fit into that. So going into a little bit more on the business side of it, the business motivation I'm going to probably go a little heavier on that. This is a data architecture strategy series and I'm sort of assuming a lot of folks can know a lot about the tech and I think, and again, I'll use myself as an example, learning more about how to sell that tech to the business or explain that tech to the business is a challenge for a lot of people and I think what turns a data management initiative into a data strategy. So I am going to go a little heavier on that and hope you're all okay with that. So we will talk tech a little bit too. So a big thing and I hope I can share my scars and battle wounds and positive experiences with you. That's kind of the point of a lot of these webinars is I would think the first thing to think of in a strategy step back and think is are we going more offense or defense and what does that mean. So are we thinking am I a startup in San Francisco and I want to make it big and it's all about profitability and revenue and competitive advantage and we're in our jeans and t-shirts and we're working until midnight and this is like awesome guys right we really want to grow probably going to those people and saying whoa we really need to think about our data retention policy now that that's not true you obviously need that but you need to think of your language of yeah we could make so much more revenue if we had a better view of our customer and that data was protected and that sort of thing right defense is what you also don't want to do is go to a healthcare organization that's worrying about HIPAA or a major financial institution that's worried about regulation or just had an audit be like guys let's look on social media data and let's just go big and go fast they're probably pretty conservative and really worrying more about risk and security and privacy right and both are needed for every single company you can't only do one that's kind of the purple in the middle but it's a spectrum and part of it is a way of talking and understanding I probably no surprise you probably speak tend to be on the left side tend to be a glass half full kind of person and I might be overdoing that when I'm talking to the auditors probably really want to think more about retention policies and roles right at the same point you just want to get that balance so some questions to yourself as you develop your strategy and or think to talk about different audiences am I going to talk to the audit department different than the sales department I hope so they're going to have different viewpoints but what spectrum is your organization what spectrum are you on and make sure you kind of just think of that accordingly otherwise I have seen strategies fall in their face because maybe you will talk to all about compliance and people saying guys we're folks and revenue that that's cool I understand but you know time for that now right so just think about that and then I saw a note a little bit of these need to be a little more concrete you don't want to be big picture absolutely again that went into that zoom out and then zoom back in that's almost like a data architecture right you have to do the conceptual model top down you have to do also a physical model bottom up and kind of meet in the middle same thing with a business strategy for those of you who are architects on the call I think you may find that approach really helpful there's a lot of business modeling tools out there and broader enterprise architecture that can work really nicely to really communicate those business needs but when we go to look zoom in on generally someone is going to talk about ROI return on investment there's most organizations even nonprofits have to live and breathe on money that's part of the world right so there's some sort of business case so generally you need to do whether it's an actual numbers ROI you need to show or a gut feel ROI again read your audience I probably made all the mistakes you will make going to a bunch of you know the finance department for major bank and not having numbers in front of you might be a risk or going to a an omission driven organization and talking too much about numbers and not talking about their mission right you need to judge that but all of the cases have these sort of aspects so decreasing costs absolutely and that's one of the easiest things to show with data management because almost every company has a lot of wasted labor costs data cleansing manually integrating that ever present spreadsheet we all know inefficient processes and I've got a colleague and he likes to say if you're not doing data quality you actually are you're just spending too much time you're not doing it well at some point you're going to be cleaning it up increasing revenue that's often we often forget to quantify but I'll talk later about getting business sponsors on your side can you can you work with someone like marketing or maybe you are from marketing on this call and say hey you know if we had better data about customers or competitors or about our product features then we could drive an extra amount increase in our campaign you know click through rate or something right could we optimize our prices through analytics you know could we there's a lot of things you can do with data which is why data is so hot right now to really help drive revenue that's another one reducing risk again think of that offense defense but gosh I had one customer their whole point for getting a governance and metadata program in was GDPR and they did the calculation that they got that GDPR fine how many multiple millions of dollars that would be that was an easy case to me right but it isn't always only only money you know product traceability we worked for the couple big food companies and understanding where that you know farm to table or where this fish was caught if there's ever a problem or where this cattle was grazed so you know a lot of different areas health and safety if you're any of the food company or or a lot of that again is sort of inherent and often can be the easiest way to make the case for change but don't also forget protecting the reputation of your company could be customer satisfaction also some examples of that but how many of us haven't had an example where you know they didn't know your customer who you were and what a bad experience and you lose trust so maybe you don't absolutely can't show a one to one decreasing cost from that but over time you will so the more you understand your customer through data that's where you're going to get your customer loyalty your stickiness especially in the day of social media are you doing social media listening through data to really understand that because that might bite you if you don't so apologies but if you've been on these you know it's coming down as daily rant about data but hopefully this rant that's sort of a collection of real-world examples might resonate with you and it might just be kind of a way to think about it so I am and so twisted having lived with data most of my life but I see everything through a data lens and my poor family sort of knows that when something happens wrong a bad customer service or they didn't get something on time they're like it's a data management problem done I told them it was a data management problem and I generally say well good luck with that I'm sure you got a great answer but so many things are a data management problem so this is kind of a super set of a lot of problems I'm still having in my life my credit card company every month for the past probably about 15 years sends me a bill and says my balance is due they've finally been able to get online after many times reminding them and they still send me limited time offer and roll in this credit card every single month so that's annoying and that's also a waste of money I am already a customer this is a classic I'm sure what everybody's mind this is a classic master data management issue one Donna Burbank is a customer with a single view of customer which means I have the product and they're also marketing to me as a customer which is a waste of time and money also I'm sure you've all all gotten these sort of things dear I've had this as example dear no name right so there was a field that wasn't populated because I didn't really don't put my name thank you for being a valued customer that went over well or I have gotten dear Joe dear Mary I've gotten the wrong name dear Donna Spelldron etc so a lot of problems in there but when we talk about being more digital what if they just automatically paid my credit they increased my credit limit because I know I've been buying a lot of airline flights or whatever right so this is sort of a real world of data-driven digital transformation right so what are all the things wrong even the simple example of just trying to get me my credit card bill every month you've wasted all that money that they still send physical paper snail mail who does that anymore right and so they've lost revenue because maybe I might have been interested in this limited time offer but pick an offer that would make sense to me like if I spent more on my credit card you'd give me a prize or something right but to say please get this credit card this is just wasted all around you've wasted money you've wasted revenue and you've lost my brand trust right I mean maybe it's not so bad because I I still have the credit card so I can complain about it right but wouldn't that be better if you really had that data-driven transformation powered by things like good master data when you truly knew me and my first friend was too creepy but understanding that recent precious history maybe we've increased your limit or here's a coupon for whatever you buy but again there is so much opportunity and I see too many companies even squandering it by not even getting those absolute basics right so hopefully that was sort of a helpful example and you can probably apply that to your own company so KPIs are key performance indicators to steal Peter Brooker's quote of you can't manage what you can't measure and I think a lot of us are used to this and data management financial KPIs have been around since the dawn of time I think even on the Pyroglyphs they have you know counting sheep or whatever they had so we're pretty and a lot of us build dashboards for budget goals expense ratios revenue projections and that's just a given that you have that for your financial data are we doing the same for our data assets we're always doing data as an asset right if it's financial if money is an asset we have no problem having the entire accounting department well we have a whole department called data governance or a whole company with data governance managing our data assets we should use the same rigor so do we know that that data is complete that it's accurate it's time when so a lot of you are sort of familiar with the data quality measures but then do we also apply how much ROI return savings this is having so if we don't work with companies that are trying to go digital they're trying to do a digital transformation they want to send text messages or emails instead of physical it's now mail that I showed and they look and they do data quality and they say we don't have emails for everybody or 16% 60% of them say me at me.com or go away at somebody else.com right and so that's going to really hamper your talent you could do a direct correlation between we want to have a digital transformation and we don't have email addresses right such an easy thing simple thing but if you don't have that it's not going to be right. So one way we look at this again kind of pulling some of these things together what's your business driver what's the KPI we want to manage and what's the business benefit is it costs again back to those categories is it cost is it brand is it innovation is it whatever and try to get some metrics but then also think simple right so I've done this myself I'm a data person I can get really carried away with all the ways you can slice and dice data quality somebody in sales just wants to know how is this going to make me money right so can you do a red amber green or easy ways to say if you do this you could get this but apply it to data quality in many cases you're going to want to do a full business case because unfortunately or fortunately part of your job is finance right and so you may just want to do a back of the envelope and then you can get some of that information with some of those gut fields that feel that that's a word and often it's helped to really start to do a full projection get finance on your side if this is new to you and you know what what are some of those costs we could reduce people time software time you know could we have the revenue benefit from some of this data that we own and really get an understanding of what what you could get from that okay so I just spent a lot of that that well from experience you don't get that right no one's going to listen to any of the rest of it and so definitely spend probably more time on that than you might have and you'll always have to remind them of that but then also do look at your fit for perfect solutions in your data architecture so I also though we tend to always have some sort of very visual or very kind of use case based example of what's operational what's analytic data what's your reference data and what's your metadata but think of that in the real world everybody is always oh we need cascading we need real-time streaming and we do you I what are really your use cases I know that's so obvious but what are the use cases and what are we using it for and even such a simple example like this can go a long way because I've seen the wrong tools used for the wrong job we're using tasks for master data we're using a spreadsheet for master data we're using an ERP for master data or we or you know we have two master data systems or whatever but sometimes just these really simple kind of graphs and what are we trying to do and what technology are we using it for can be really really helpful especially for non-technical people we're trying to say what's reference data well you know when we have that form and the people's addresses are always wrong if we had a drop-down list of state codes that would just go really far and you know just give examples when you're talking to the business people on my my rant I once registered for a data quality webinar and they had free text fields for all of the city and state codes and maybe that's a data person joke but that's like the worst way to ensure data quality errors right so think of those types of things and then do look at the different types of technology out there and use them appropriately so this is again from a survey we did we do them yearly now and to kind of say what platforms are people are using probably no surprise relational database in terms of current news still leads the top people love to hate the relational database it's absolutely a great tool for what it does which is keep your data clean and keep it integrated does not mean that's your only tool in the toolbox right what's the excuse me up at night is that high level of spreadsheets spreadsheets are fine but not for massive data integration and so there's a lot of different types of things as well and most companies are using a multiple platform fit for purpose so maybe your your finance data is the operational data is in relational but maybe we want to put a graph database on top of that to see fraud patterns or something right so yes but not everything goes into graph right vendors love to do that our tool is the only one you ever need but listen to them you know if you're building a house you have a lot of different tools in your toolbox not just one right what's also interesting is going into the future relational databases are still hot they're not going to go away they're a great tool for what they do you will see a bigger shift to the cloud I think Dan mentioned that in the beginning as well not a surprise it makes a lot of sense for the right use case again not the only tool in the toolbox doesn't mean everything needs to go to the cloud but you will see what makes me feel a little better those other categories are all getting a little higher because people are discovering that there are tools not everything needs to be relational as well so if you haven't looked at some of these other technologies to take a look in your strategy is there something we are missing there's a lot of really hot stuff out there now that may not have been the case when you were sort of growing up with with data management whatever even if you grew up with data management last year right there's a lot of different tools and look at them all for the right use case so the thing that does keep me up at night is that sprays I'm surprised when they say what is your future platform that 35% of people admitted to using spreadsheets if anyone knows you shouldn't so the fact that people are planning for that to be the reality goes keep me up at night but we will keep trying so again this can all be daunting yes I need to look at all these different technologies and align them with the business case and everything but again when about zoom out and it's one of the tools we use and I've had some really good success in just our two workshop get both business and tech together and you may have used these before there's different names for them but they basically what are the benefits we're going to get what's the difficulty and just sticky note it right and you I've found that you get to a fairly good degree of consensus fairly quickly this way so right what you want to do is the stuff in the upper left the green it's really high benefit and it's really not that hard can we get an address validation tool out there put it in and then our data quality increases that might be a quick win right customer master data is super beneficial but it's also pretty high difficulty right doesn't mean you're not going to do it but it might help with your prioritization what you don't want to do is stuff that's really hard but not very beneficial you can read the chart but sometimes these low benefit low difficulty you know maybe it's not a bad idea to get done so this is just a nice way with with both business and tech to really help do that or I guess an Agile is that that t-shirt sizing right with the benefit and one of the big ticket items and it's a nice way to get to consensus fairly quickly if you go too much further the other thing to think of and this is and can be an entire webinar ends up itself so I'll just touch on it but it probably deserves 40 slides of its own is thinking of organizational capability and how your governance structures really fit into that so the people and the process are as big if not more so than all of the tech so I kind of waited this webinar a lot into business drivers because if you don't do that nothing is going to work and then I went into tech and but don't forget the people side because that's really what's going to help and think of how your organization runs is it more Agile is it more federated is it very top-down so really think of that when you're designing your strategy of who is going to implement it and what are their roles both business and tech and then I'll say it I'll say it again I'll say it again just like a good marketer don't forget the marketing piece put it put a branding campaign your strategy really hit that home have lunch and learns have t-shirts have stickers on the laptops whatever whatever is going to kind of help make things saying and plan ahead and as you're finding trying to get the buy-in find advocates across the organization that aren't you so if you are a technical person find someone in marketing to go for the to the board with you find someone in you know patient advocacy or something that isn't you that helps to say that yes we're body into having data be part of the business this is going to go a lot more way further way and and and another piece so kind of bringing it all together is how do you bring that into an actionable road not again a bit of an art a bit of a science but what I would just say having done a lot of these you want to do a quick win but don't do a quick win that's going to be a throw away your quick win could be data quality it could be a date conceptual data model is often really popular but try to look at it holistically so what what's the big driver for the business maybe it's integrated customer view because we're going to do a new product launch and a marketing campaign who cares who are the key stakeholders marketing sales execs right and then what and when so there may be a campaign coming up try to align with that and then what do some foundational things like a business glossary to be really popular but that's architectural and some sexy things do some customer analytics right don't only go with old school it has to be a mix hard stuff easy stuff shiny stuff and then time it around things that people care about so that's a lot I know and I talk fast so I'm also a big fan of that again when in doubt just think of those who what where why when kind of the Zachman framework of data strategy and I've gotten some good feedback that this is a really nice way that people have used to kind of just again whiteboard their strategy why are we doing this think we cover that a lot who are the stakeholders who's the stewards who's our executive champion the how and just think of the how is it only a technical how not only how the data is stored but how is the governance implemented who are the people and then the what of course what data what platforms real-time batch etc and then the when how do we time all those together it's going to align with the needs of the business so in summary this can be a lot and but these can be broken out into small chunks that are lining the data in the business and link all that together with the people process tech but try to get those quick wins so as Shannon opens it up for questions because I've seen people being very active as usual which is great I'll do a quick plug for we do this for living if you need help let us know another quick plug for the white paper which is able to be downloaded and a third quick plug for these are the webinars coming out I do want to give a plug next month because we have a real-life case study which we're really excited about from which is a big construction company that did some really great things with data modeling so you'll want to join that and without further ado I will open up the Q&A Shannon amazing thank you so much Donna for this great presentation I think today I'm welcome to join in the Q&A here I'm going to dive in is the data strategy the same thing as the data management strategy the data strategy I think touched on at the beginning I would say is very is a super set of your data management strategy so definitely think of all of those things in the day with the talk that you need to manage but then think strategically about it so add that business vision the why the what and the who around it to really make it more strategic and less just management that I know if Dan has thoughts on that as well no I totally agree with that perspective Donna great we determine where to start yeah so I think I'll keep saying it enough that that marketing paradigm that if you hear it six times you might remember it when you're brushing your teeth right that when in doubt zoom out so think of that who what where why when and then some of those tools that hopefully have shown you to start to whiteboard it just to just do on a piece of paper or whiteboard why we're doing this what are the business drivers try to map out a very high level architecture try to think of what other projects we can align with that's probably the best way to start in the who don't do this yourself maybe it's a personal thing we want it to be our strategy that will never work you want to find what other champions it's the best way to have your strategy that you don't if you're a tech making the assumption here a lot of you are tech maybe you shouldn't present maybe it's somebody else in the organization that really champions your strategy so it has to be an awesome so this is a nice way if you're if you're feeling completely overwhelmed with it just start to ask some of these questions and then just start doing and start kind of planning it out without getting just overwhelmed anything you want to add to that yeah I would say that we always advise go for the quick win show momentum as quickly as possible and and Donna you had a funny phrasing it was the shiny things don't forget about the things you know particularly analytics helps you do that you know customer insights or other things that you can show some some quick insights that have big impacts you know don't always do the back back office stuff first go for the shiny and quick and that'll help build momentum and champions internally especially on the business side yeah no I agree with that you have to show some quick win the people get it and see some progress it is an only analytics sometimes I thought one of the comments it could be quality or sometimes showing the analytics is the best way to show the quality human nature right you can say you know our our customer is terrible it's terrible okay whatever and you show them a report and they come back to you are the quality in report is terrible but you have to do it some way that people can start to put their brains around and analytics is a good way to do that because it's something tangible that people can see it isn't the only option but it is a nice thing that people can kind of get in their hands to sort of at least see where you're headed do you have to provide an example of a good data strategy initiative and the level of detail that is included um oh gosh there's a lot out there I actually see the one slide the slide they remove are the ones that come up I had a slide on what is it what is the data strategy there's a lot published you can just Google it a lot of governments now are having their data strategy my bias I would say I have a lot when I give a lot of seminars and workshops on this a lot of people come up after it's like but what is it like what's the deliverable is it a big document and I would say if you're doing a big document you're probably on the right on track sometimes yes that has to be available I would start with a PowerPoint I would probably sell that message and what's the plan is probably the best way to start because that's going to help you get the buy-in and it's going to help add that clarity of your thinking and I think the best data strategies are the ones that are working it's Amazon.com it's Uber it's probably not the big thousand page document that you do find when you Google because someone spent a lot of time but is it being action so I think the best strategy you probably don't even notice is the Amazon.com recommendation engine because you're using that but I'm again yeah we tend to get involved after the data strategy has been defined and then they try to figure out how do I implement this so we fall down the next step to that and that's kind of what I showed in the reference architecture for modern integration and automation. One general comment though is that once your strategy has been defined thinking about a data ops approach where you bring together IT and business and make it really iterative as you start to deliver data as you start to deliver business value to analytics or new applications how you keep those cycles as short and iterative as possible and how you can respond very quickly if the data model may be wrong the analytics report may be wrong you may need to enrich or add data that's something that you should think about as well. All right well thank you both so much but I'm afraid that is all the time we have for today's session just a reminder that I will be sending a follow-up email by end of day Monday with links to the slides and links to the recording for everybody as well as links to the research paper as requested again. Thanks so much everyone for being so engaged in everything we do and thanks to CLIC for sponsoring today I hope everybody has a great day and stay safe out there. Thanks everyone. Thanks Donna.