 Hello and welcome. My name is Shannon Kemp and I'm the Executive Editor for DataVercity. We'd like to thank you for joining today's DataVercity webinar, a data-centric strategy and roadmap supercharging your business. The latest installment in a monthly series called DataEd Online with Dr. Peter Akin brought to you in partnership with Data Blueprint. Just a couple of points to get us started. Due to the large number of people that attend these sessions, you will be muted during the webinar. If you'd like to chat with us or with each other, we certainly encourage you to do so. You can click on the icon in the upper right corner for that feature. For questions, we'll be collecting them via the Q&A in the bottom right-hand corner of your screen. Or, if you'd like to tweet, we encourage you to share highlights or questions via Twitter using hashtag dataed. To answer the most commonly asked questions, as always, we will send a follow-up email to all registrants within two business days containing links to the slides. And, yes, we are recording and will likewise send a link to the recording of this session as well as any additional information requested throughout the webinar. Now, let me introduce to you our speaker for today. Peter Akin is an internationally recognized data management thought leader. Many of you already know him or have seen him at conferences worldwide, especially the Diversity Conferences. He has more than 30 years of experience and has received many awards for his outstanding contributions to the profession. Peter is also the founding director of Data Blueprint. He has written dozens of articles and eight books. The most recent is Monetizing Data Management. Or has that changed yet, Peter? Have we published a new book yet? We turned it in. Awesome. I don't know if we can actually change the numbers, but soon. Awesome. I love it. Well, and Peter has experienced more than 500 data management practices in 20 countries and is consistently named as a top data management expert. Some of the most important and largest organizations in the world have sought out his and Data Blueprint's expertise. Peter has spent multi-year immersions with groups as diverse as the U.S. Department of Defense, Deutsche Bank, Nokia, Wells Fargo, the Commonwealth of Virginia, and Walmart. He appears at conferences often and is constantly traveling. Peter, where are you today? So, Shannon, I am talking to you from the middle of Empire State Plaza, which has all been in New York. And if you haven't been here to see this architectural feat, you need to come and visit it one time during your life. It is a phenomenal feat. I have to tell you one thing, though. They do have a naming problem. This thing in the bottom right-hand corner there, my wife says, that's not an egg that looks like something from the Jephins. Well, okay, maybe it is. But anyway, I've got a live group here today. What is it? A deviled egg. There we go, or half a deviled egg. It's actually a concert theater, so very, very cool place on this. But thanks, Shannon. It's good to talk to everybody today. First webinar of the new year. And as you said, we're getting ready to turn over the ticker from nine books to 10 books on this. We turned in the data strategy book last week, and we're waiting for the editor to come back to give us, you know, those feedback where they say, oh, you're not quite done yet, or maybe you're done. Who knows? We'll see. But anyway, we're going to talk about this topic today, and dive into it in a little bit more detail. Specifically, we'll talk about three things to get started. First of all is a statement. A data strategy specifies how data assets are to be used to support the organizational strategy. So then we'll talk specifically about what is data, what is strategy, what is the data strategy, and how do they work together. The second statement is a data strategy is necessary for effective data governance. You're going to use the data strategy to improve your organization's data. You're going to use the data strategy to improve the way your people use their data. And lastly, you're going to improve how people use your organization's data to support their organizational strategy. Third part, then, we're going to talk about effective data strategy prerequisites, which is that there is a lack of organizational readiness. We need to compensate for lack of data competencies all around and eliminate the barriers to leveraging data. We talked specifically a little bit about the Seven Deadly Sins, but we went through the Seven Deadly Sins on the last webinar of last year in December. Also, I'm just going to briefly touch on this. The other aspect of this, then, is how you do data strategy in terms of iterations. We're going to dive through that process, but it basically comes down to lather, rinse, and repeat, which most people laugh about, but actually, the chemical companies have pretty much got us there, right, because we never get out of that loop. And that's what your data strategy should be about, as well. It does require a balanced approach, however, and then we'll get to the best part, which is the Q&A session. So let's dive in, take a look at what's going on here. First thing I like to do is to point out to people if you haven't had a chance to go out to see Simon Sinek's great TED Talk reference in the upper right-hand corner there, how great leaders inspire action you really should, because what he does is he makes a very simple point, which is to say that many people are very good at telling you what they do. Fewer people, however, are good at telling you how they do what they do, and fewer still people are good at telling you why they do what they do. And his point of his whole 10-minute TED Talk there, although I would urge you to go out and watch it, is that it's not what you do, it's why you do it, and that's an important distinction. People don't buy what you do, they buy why you do it. The contrast point that I like and that he makes in the video very, very well, can you imagine how less impactful Martin Luther King would have been if he said, instead of, I have a dream, I have a plan? It just doesn't quite work, does it? So it does get to the difference between the why and the how and the what, and that's what we want you to think about from a strategy perspective. Strategy, of course, is really derived from a military context. We've used it mostly since the 1950s, and it got up to fairly good peak usage in around 2010. It's actually been falling since, so it's one of those wonderful things on Google Trends you can actually watch how the word is used. My favorite definition for strategy, however, is a pattern in a stream of decisions. Now let's take a look at how that might work in a particular example. Army troops for Napoleon are not necessarily going to have a lot of time to think when they're in the heat of battle. So Napoleon had to come up with a strategy for defeating the British and Prussian troops. You'll see it there on the screen. I'm going to show you a close-up to this in just a second. But the Prussian troops are in black on the right, Napoleon's troops are in blue, and the British troops are in red. Now what Napoleon understood that perhaps his enemies didn't understand as well is that when you hit an army and force them to retreat, they tend to retreat along their supply line. And if you watch, this is a perfect example of Napoleon's strategy of dividing and conquer. What happens here, of course, is that we look at Napoleon's army, which is set up right here in the middle of the blue, and it's going to hit the red and the black very hard, just like trying to bowl a strike and push the two of them apart. When they have to move, they're going to move towards their food, just a natural thing to do. And his strategy was first we then go beat the Prussians and then we go beat the British. So it's a three-part strategy. Hit them really hard, then get the black, then get the red, or die. That or die part puts a little bit of incentive around the whole process in order to do it. Strategy's got to be that simple because if strategy isn't that simple, then we are going to have a problem in terms of getting people to follow it because communication is hard enough as it is. I have to make a slight note of it here as well. Diane says I've got the wrong color on Wayne Gretzky here. The proper color should be Diane. Blue, white, and red. Okay, blue, right, and red. Apologies for not localizing my presentation here for the audience here, but meanwhile you're saying what the heck is this? Well, Wayne Gretzky also had a very simple definition of strategy. He didn't skate around chasing the puck because he would never catch it. He skated to where the puck was going to be. Strategy has got to be something so simple, so easy to communicate that everybody in the organization can follow it or else. I'll give you one more example of strategy from the book that we use in here at Walmart. Everybody's familiar with Walmart. Their strategy for many years is very simple. Every day, low price. Everybody knows that. Every associate, every supplier, every customer knows at that point in time. Now as data people, we have to come back and look at data and say, well first of all, we think data is special compared to other assets. Data is the only asset that we have in our organization that is non-depletable. It doesn't degrade over time. It is durable in nature. When I say durable in nature, that's an accounting term, but what it really means is something that we invest in as opposed to something that we expense. So it's treated differently on the balance sheet. No, we have not managed to put data on the balance sheet. We are working on that very hard. It's a long, tough problem and we have a long way to go before we get there. But as far as strategic assets go, it is right up there with your other strategic assets. In fact, when you look at data assets compared to other assets, data assets really do win in the sense that they cannot be depleted. They cannot degrade over time. They are durable in nature as far as compared to the other strategic assets that you have. However, I'm working on a blog posting now that says data is not the new oil. I was in the Middle East earlier this year and it was a lot of fun to work with them on this, but they were very disappointed because they thought they could replace their oil supplies with data supplies. The reason data should not be thought of as the new oil is that you think of oil as a production function and once you use it, it's used up and that's the end of it. Data, of course, can't be depleted, which is the main difference it has from there. So I like to add the letter F in front of it that the data is the new soil rather than the new oil. Plant something in it and good things will start to happen. If you need to call up the new bacon, that's fine. Actually, I changed that to chocolate in the Middle East just because a new bacon wouldn't be a terribly popular metaphor there. So collectively, what we're all trying to do, of course, is to help the organization that we work for unlock the business value of its data by strengthening the data manager capabilities that they already have, providing solutions that are appropriate to the organization, meeting the organization on the journey it is on as opposed to trying to... I'll use the analogy a little bit later on, but give a 16-year-old the ease to the Tesla and really building lasting partnerships all the way around that as well. So let's dive on a little bit further. When we talk about assets from an organizational perspective, what we're really looking at here is the idea that data should also be included in this list. Let's take a look at that list. Better Shannon, we'll hopefully get that taken care of. Data should be on that list of assets. And one of the easiest sales I ever have to doing this is that most of the HR people kind of get it and go, oh, yeah, like people are assets, data is an asset, too. We really need to think about them kind of the same. And I say, yeah, that's actually pretty good because the idea behind managing your people is you've got some HR managers that take care of those people. So we should have some data managers taking care of the data assets, too. Oh, wow. CEOs are getting better about this. 33% of them say they're actually measuring the type of information assets that they have. Quarter of them are quantifying the value that's produced out of it. A fifth of them are cataloging at least their assets. 10% of them are not paying any attention to it at all. And 10% are saying they directly monetize it, which is kind of useful around there. I do have another book out on that monetizing where I'm going to dive into that today. It's another webinar entirely. What we're trying to do is get people to think of data not as this thing that's squished between IT and the business, but that really we're going to look at it from a much different perspective, which is to say that data really is the thing that unites the business and the data and puts it in between the two of those. You can always say, and I'll quote Michelin Casey all the time on this, there will never be any data, less data than right now. So let's go back to data strategy. Three functions of your organization's data strategy. It should improve your organization's data. After all, if your data gets worse with your strategy, you've got a problem, right? It should improve the way your people use your data. So your data strategy can't be focused circling on your data. It's got to be on how people use the data. And then finally, improving the way people, your people use your improved quality data with their new improved knowledge about how to use data in support of your strategy. Now if you're in a nonprofit organization, substitute the word mission there. It works. There is no problem with having these things go back and forth on that. So a data strategy is the highest level data guidance available in an organization. It focuses on data-related activities on articulated data goal achievements and providing directional but specific guidance when faced with a stream of decisions or uncertainties about organizational data assets and their application towards business objectives. When Walmart thinks about data, they think every day low price, what's the right decision about my data? When Wayne Greski is looking at the puck going around the rink, he thinks, where is it going to be next so I can get it and score? Napoleon didn't have quite as much data, but he at least knew where the supply lines of the troops were, and he forced his group to go into it and back up. Let's take a look at how data strategy and data governance then work together. Data governance has been a huge topic. Data diversity runs wonderful conferences on it. We're going to be down in San Diego in June. Thank you for that. I'm going to talk about some of these ideas with everybody else. It's a wonderful event. If you haven't had a chance to get there, please do come. It's a really, really super time. The hotel is funky too. Anybody that hasn't been to the catamaran knows what I'm talking about. Great place to do it though. We have a wonderful time, and it's a really good exchange of ideas. Data strategy then says what the data assets do to support the strategy. Data governance then says how well is the data strategy working? These are the two feedback loops that are incorporated in there. What we found is most people really weren't aware of how that worked. Let's put it in a little bit larger context. The data strategy is the data portion of the organizational strategy. The organizational strategy might be hire a bunch of engineers. If you discover that your company has, you know, all of your engineers are at retirement age and you'd better hire some new ones. That's probably a tactic at that point, but nevertheless it is something we can talk about from a strategic perspective. The data governance then should be talking about how data is delivered by IT. And those IT projects then go into organizational operations. We certainly want to put a feedback loop into that process, as well as acknowledge that there are other aspects of organizational strategies that are outside of the data strategy. Although I have seen a lot of writers that say your data strategy is the only strategy the organization should follow, they found an organization that could balance their entire strategy on just the data by itself. Let's go a little further. This is something that we came up with called a data sandwich. It's kind of inelegant at first. We've got this concept out there called data literacy. People are kind of, what does it mean to be data literate? If you Google the word, you'll find a lot of different phrases. We've got, I think, a pretty good definition in the book, which has got some very specific pieces. We'll go into this one when we get to the QA section. Then, of course, you've got your data supply. And your data supply is kind of uneven. It's probably also largely filled with rot, data that is redundant, obsolete, and trivial. And then you have some sort of vague idea of what standards are in the middle of it. So that's not an appealing sandwich, at least to an engineer like myself. So when we talk about what does it mean for your organization to get data literate, we're talking about how can we make people understand the data better, understand how to use data better, and then understand how to use that better data in support of your organization. When we talk about cleaning up the supply of data, we're talking about two things, getting rid of the extra stuff that we don't need, and making the stuff that's left over much, much better. Our research shows that 80% of organizational data is redundant, obsolete, or trivial, which means it really just gets in the way. Now, I always astound people by asking them how many sources of customer information they have as we're looking around, and most people just say you don't want to know. And I say, well, the average is 17, and a couple of them go, wow, we've only got 12. We're above average. That's actually kind of good. What we do, of course, also want to do that is work towards standard data. This is not to say that we're going to standardize everything, but the more things that we do standardize, the easier our task will be. And notice that our data sandwich now has gotten a little bit neater, at least in my mind as an engineer. That is a much more palatable data sandwich. One of the reasons I push these things out here through the webinar to you guys is because this is probably not the best way to do it. And I'm certain I'm going to get a call or a suggestion from one of you guys on, yeah, sandwiches are really terrible in allergy. Do something different. Maybe it's a hot dog bun or something like that, or a canoe, or I don't know what. But come up with something that works. We'll share it with everybody else. So that's how data strategy works together. It works with your organizational strategy. If your organizational strategy is not good, your data strategy is going to have a hard time doing something else. And you want to make sure that the two of them are as complementary as possible. We've already seen that there's a reference check between the two of those. Let's talk about what people have really had trouble, which is that data science hasn't worked so well for everybody. Ten years ago, people were talking about it because it was a job of the 21st century. But my colleague Eric Siebel said greatly, I wish I'd said this first, but he did it first so he gets credit for data science. If they redundant term, it's like saying a book librarian. And data scientists are only 20% productive. All to 100% say they spend 80% of their time munging the data, which is what we're trying to talk about in terms of how we do all of this. So I only got to get them from 80% unproductive to 60% unproductive. I've actually doubled their productivity, which is a wonderful thing, but still a very sad piece. I wonder how many of you knew we were in the post-big data era already? Yes, we've done that. It's all these Vs that we talk about, and it's really not terribly interesting when we look at them because they are not objective in nature. I do like this particular saying that because the last V on here is the vanity of the big data experts. Because we have no objective definition of big data, it's very, very difficult to describe any particular benefits today. We don't want to have it be a subjective determination. We'd rather have it be an objective determination on this. So don't talk about big data, instead talk about what we can objectively identify, which are big data techniques and big data technologies. Now, let's get back into this. When we're talking about the data strategy and how the data strategy supports governance, again, it's what the data assets do to support the organizational strategy, which means your data strategy has to be phrased in terms of business goals. I need to make this number go up. I need to make this number go down. If I can't, I have a different set of problems. And the data governance feedback has to be in terms of metadata. Now, those of you that have taken my data governance seminars know that I absolutely insist that the data governance people speak in the language of metadata, because if you don't speak in the language of metadata, it becomes very difficult to figure out what you are doing and how it affects the business. Let's talk specifically now about how data strategy should be put in place. What are the reasons for it? Again, improving your organization's data, because data points to where valuable things are located in your organization. If it's a source of expertise, if it's a source of money, if it's a bank account for one of your customers, it's information that they've given the government three or four different times. This is all valuable stuff that people have that they're exchanging with us in return for goods and services. Data also has an intrinsic value by itself. When you look at data, there's an awful lot of things that it can do. Just knowing that the temperature in Virginia this week with six degrees at the beginning of the week talks about how people are going to approach their day. We started by closing the universities because we can't deal with six-degree weather, unlike up here where they say, ah, six degrees is warm, right? Well, not exactly. It's somewhat balmy. And finally, data has an inherent combinational value. If I'm L.L. Bean, maybe I'm looking at the forecast. Maybe it's even something as traditional as the Falmer's Almanac and saying, it's going to be a colder weather let's bump up the supply of coats by 10%. Very simple type of measures like this, but if you take the different piles of data that you have, the different data assets that you have, and combine them correctly, and again, that assumes that you can identify where they are and that you understand what it is that they're describing, then you can get to this third layer. Notice each of these is an additive function. If you don't know what data points to where things are located, you can't combine it with anything else. If you don't know the data's intrinsic value, again, it's not going to be helpful as well. So all of these things are additives in nature. Again, improving the way your people use its data, because people are generally data illiterate, and I don't mean that badly, but they don't understand the function of data. We've had an awful lot of talk after the election this last cycle through about fake news and whether people understood certain things were true or false about the various positions everybody was taking. If you don't have the data literacy, you have a big issue. For example, I remember there was a blog post at one point that said that somebody was going to deport 300 million illegal immigrants in three years. You say that's 100 million immigrants a year, which you divided up as about 100 million immigrants a year, divided by 365. Well, that's like 10,000 immigrants a day. How could we actually get a program in shape that was going to do something like that? The physical impossibility of it is just overwhelming when you start to break down the facts, whether or not you're agreed on the general position or not the facts are the one thing that we have to understand and we have to use data to measure the change. Are people getting more data illiterate? If you say, yeah, sort of, it's kind of like coming home at the end of the day and saying, hi, how was your day? My day was fine. How was your day? No information is changing hands when you do that. You need good, hard data to measure the change. You need to use data to manage the change because if you don't use data to manage the change, you're not going to be able to keep people going in the right direction. If you're steering the ship and somebody says we need more coats at LL Bean because it's a cold winter, do you double production? Probably not. By the way, I know nothing about LL Bean, so I'm just making this up on the spot as far as the example goes, but somebody's going to say, okay, have we made enough coats? Are we meeting demand? There's got to be some data that we can use to do that. And finally, we need to have data to motivate change. Now, I do have an example of this because motivating change with people is pretty easy. You just affect the things that they need to measure. So one of the things I'm pretty sure that Amazon doesn't want me as a prime customer to do is to order everything in single units. I have Amazon Prime. I live way, way out in the country. It's a one-mile trip to my mailbox and a 10-mile round trip to get milk if you forget it. So when we get out there, it's pretty far out. And I ordered 219 individual Prime orders from Amazon last year. I'm pretty sure that wasn't what they intended for me to do. But what I do with my orders at Amazon is I order them on the Prime, but then not I don't need it in two days. And if you get that, why they give you a dollar off of music? And as you guys know, we like music channel 9, so we're always playing something at the beginning of the webinars here. And so I downloaded $219 worth of music songs. Things last year. It was fun. Well, that's me being AAD, first of all. So seriously, be careful on that. But I can guarantee you Amazon did not put Prime in place so that I would take every order that I have instead of batching it into one thing. I order each one individually because each time I order a don't ask for a two-day delivery, I get a dollar in download music. Boy, what a nerd, right? I mean, this is really serious stuff, but it's how you motivate change. Now, it's not the change that Amazon probably wanted, but at least it worked out for me. And of course, I'm advertising here free for Amazon, so I guess that's probably good for them too. Finally, of course, we look at the way your data and your people support the organization in there. And I'm going to give you a very quick example on this just to show one organization the way they did it. I showed this example earlier. Sorry, guys, you've seen this once already. But the example is Rolls Royce, which makes really fine jet engines and gets attached to things. And just for the record, we have some GE people here today too, so they make really good jet engines too. I don't want to leave anybody out, but I got the Rolls Royce commercial before I got the other one set up in here. So the old model of doing jet engines was that Rolls Royce sold them to the various airlines. And when they did, they sold the engine to the airline and the airline paid attention to jet engines, but they also paid attention to meals for people, seat belts, paint, pilot training, tires on the airplane. As many things to look at in an airplane as there are items in a Walmart store. Running an airline is an incredibly complex business and making sure everybody is safe and healthy is another important part of that. So Rolls Royce said we need to get better attention to the engines because our customers are underutilizing jet engines and we want them to get better at it because then we will be better as a company as well. And we can also both collectively make more money. How are we going to do that? Well, the new model is instead of selling you the jet engine, they now sell you hours of powered thrust. The first thing that's happened is Rolls Royce is now on the exact same side of the table as American Airlines because if it's not making money for American Airlines, it's not making money for Rolls Royce either. There is no payment for downtime, so that becomes now the task of minimizing the downtime in this particular process. Then we ask the question, what does this have to do with NASCAR? Now that was kind of an interesting piece to look at. So they looked at this process called Wing-to-Wing. It's going to be a little bit loud. Praise yourself. Oh my goodness, the music is going to still hang on. There we go. Let me go back to it. It caught me, Shannon. Take a minute to get back up there. But anyway, what happens is when we're looking at this hours of powered thrust thing, the Wing-to-Wing process becomes the process by which you can change the engine out. Because if the engine isn't making money for your American Airlines, it's not making money for NASCAR. But Harlan comes in for a pit stop. Time to refuel and change tires. Lewmore himself changes the tires. Only four crew members, including the driver, are allowed to work on the car. It's the tenth time. Harlan stays in his seat, anxious to get away. Let's watch. Now I'll shorten this a little bit. It's 64 seconds to change the tires. It's approximately four seconds to change the tires there. So what you're seeing here is NASCAR helping Rolls-Royce become better working with the airlines to make sure the air parts, the aircraft are back up in the air and earning money for both companies much more quickly, much faster than everybody else. So let's move on to the second half now and talk about what sort of prerequisites you need to have in order to put your day to strategy in place. Sorry, I'm going to go back to that. I want to make sure I talk about them. First one is that most organizations are really quite unready for that. One of the things you've noticed at that last example there of changing the tires on a Formula One race car in four seconds is it requires a lot of teamwork and coordination and special skills. Most organizations aren't ready for that and I'm afraid that those of us in the college and university group have to take the blame for that because we've done an extremely poor job of preparing your future employees to help you out with your data problems. I'll tell you why in just a minute. Second issue is that there is a lack of data competencies within the organizations. So organizations have really thought about data as sort of this thing that falls in the middle. Business thinks it's an IT problem and IT thinks it's a business problem. In most cases when I see this occur, people say to go on the IT side, well if they can sign on to the server, they are good and my job is done. And of course that talks nothing at all about what's actually inside the system or what's going on. And I said the last part is the seven deadly data sins. We'll talk about those just a little bit. Here's how we line them up though. Again, what the data assets do to support strategy is the data strategy and there's two phases, the prerequisites. We'll do that now and then we'll do the iterations at the end here. First one is to prepare for dynamic change and determine how to do the work. Second is to recruit a qualified, knowledgeable enterprise data executive and other qualified talent. And the third one is to eliminate those seven deadly sins that I talked about last webinar because we don't teach people much at all about how to do data. Had a wonderful collegial lunch today talking with a fellow professor who was talking about what he did with his class and how he got them to be more data literate in class. But the sad fact of it is even if you've got great professors that are doing a wonderful job there, they get one course in data, period. If you get a good person that does that, that's really lucky. If you get a bad person, it's a problem. We had conversations around that as well. And there are quite a number of people that are not able to clearly articulate the value of normalization and other topics along those lines, data integrity, et cetera, et cetera. More importantly though, management gets the impression because they've gone through the same IT courses. And they kind of get this idea that, yeah, school taught me the only time I ever should say the word data is when I'm building a new database. Which means that if I'm migrating databases, I'm not building a new one. And so I don't need to have data, knowledge, skills, and abilities. And if I'm implementing a new software package, I don't need those knowledge, skills, and abilities because I'm not building a new database. And if I'm installing an ERP, I'm not creating a new database and therefore I don't need these data management, knowledge, skills, and abilities. Now let me just wait on here for just a second because some people say, but doesn't the new software package come with a database? Yes, it does. But we're not building a new one. We're running a script that puts the data at DDL out there so that the pre-formatted database comes out. And nobody thinks twice about whether that data model supports the organizational strategy or not. I've been in many, many organizations where the basic model of the package, whether it's an ERP or a software package, does not support the mission. In fact, it runs against the mission. And when you show them this, they go, wow, I never thought about that before because it's package software and package software is always cheaper than my new software, right? No, no, no, it's really not. It doesn't mean that it's a very easy decision all the way around, but it is a problem because we haven't taught IT professionals much about this. Whether they are in positions of leadership or just in business in general or on the IT side, they don't know that they don't know. And that's really what the problem is. In fact, my co-author on the book and I have actually compared notes in over our 30-year careers. We have never been interviewed by an interview panel that was qualified to interview us. Now, this is particularly important. We did some work around this on the federal government sector where the federal government is actually leading industry right at the moment in terms of hiring CDS. It's a really good move. By the way, my favorite term for this is not chief data officer. I like the term enterprise data executive, but that's a different topic. We won't go into that there. Maybe somebody can ask a question if you're interested at the end. These hiring panels that they're doing in the federal government are asking people questions that they don't know the answer to themselves. So in other words, if you don't know anything about data, how are you supposed to hire somebody and determine whether they are qualified to know about data? It just doesn't work. So what we've suggested for the federal government and we may be able to implement this on the CDO council that we're working on right now is having the federal government share a hiring panel so that EPA can borrow the one that worked for energy that also worked for other places in here. Again, I hope this makes sense because if you don't know that you don't know, then any answer is right. I've been to probably over 50 CDO events in the past five years and half of the CDOs that are up there are CDOs in name only. They just don't know what they're doing, which you're like to spell CDO because they think it looks like a great career. And nothing wrong with any of that. We like to have people doing this. Most of you have seen this Maslow piece before. It just means that you can't get to the next level if you haven't got the first level actuated. And of course, in the basic knowledge of data, most people don't realize that in order to do the things in the golden triangle, you need to do the things that are foundational in practice which are capability-oriented rather than technology-oriented because those will help you by providing a stronger foundation. Remember, your foundation is only as strong as the weakest link because everything else will take longer, cost more, deliver less, and present greater risk than if you try to do it directly on your own. I mentioned the seven deadly sins. We went through them in detail on the last webinar, so I'll just briefly hit them here for us. Organizations don't understand what it is to be data-centric. And the reason we focused on this last time was to introduce the website that we call the Data Doctrine. And if we'd like to get you to join us, if you'd like to see what that's all about and learn how we can participate collectively in making everybody else understand what it is to think data-centrically on that. There is a lack of data leadership out there. There has to be because there aren't many data positions. Therefore, people haven't had a chance to get good at this. If you think about it, it was hard to be a CIO before computers existed. We have the same sort of relative immaturity around that process. But we don't have in most organizations a robust programmatic means of developing shared data. If you think about it, those words, robust programmatic, are almost directly at odds with the project orientation that exists in most organizations implementing their data pieces. It is a huge challenge, and we have to work on it at a lot of different levels. Once you do, you now have a data program, and that data program needs to be correctly aligned with your IT projects. If you don't, you'll have an IT project that is misaligned with your data that can't result in good outcomes. There's some other stuff in here, too, that we have to pay attention to. Most importantly, expectations. I have seen people hire, I'll give you a very specific story, the typical thing that's on most CDO agendas is one that starts out and says, okay, now you're going to need to go do an inventory of your data. In 30 years, I have never seen anybody complete inventory of their data. Where are you going to stop after all? This is a very long process. So rather than, of course, people say, how long is it going to take you to do an inventory of your data? Can you be done by next week? It's going to take you three months. As I said, I've never seen anybody finish it, period. So the idea that somebody would be held to an expectation of completing an inventory of their data, good luck. It was much easier in the old days when we only had one computer and we could take all the data sets that were on there. On my little MacBook here, I've got 3,500,000 different files. I have no idea what most of them do, either. And I'm not very good with Unix, either. Anyway, this expectation thing is really problematic. Another story that has always been a favorite one over the years is I had a very clever data manager that I worked with for a number of years. And she said to me, it's okay. I'm the data manager. I work for the CIO. The CIO has told me I don't have to deliver results for five years. I have five years to get this program up and running. And I said, did you forget that the word CIO used to mean career is over? The average CIO is only in the job for two and a half years to four years. And the likelihood that yours is going to survive that period is small and sure enough, within two years, the CIO was replaced. And the new CIO turned around and said, what? You're going to take three more years? She said, well, I had five, you know? They don't have patience for that. They're not bad people, but you have to manage your expectations. I'll show you how to do that just a little bit as well. You need to pay attention to sequencing your data strategy. It is absolutely critical that you not try to do too much at once and that you not try to do things that don't match. Now, the sequencing part, I had another diagram. I'll give you an example just a little bit on that. But the sequencing part is really critical because strategy is about choice. And it's about doing things and deciding not to do other things. And remember, Napoleon decided that he was going to hit them both first, try to knock them apart like a bowling split, and then get, critically, the Prussians first and then the British. All of this, of course, upsets people's apple carts, rice bowls, moves their cheese, whatever metaphor that you want to do. And if you don't bring in experts to help you manage that change process, you will absolutely fail. Those are the seven deadly sins. While we were doing the book, we discovered one thing that was really interesting. I've just outlined a huge number of prerequisites there. And the idea that anybody is going to walk into an organization with all of those talents and abilities and be able to actually do data stuff as well, that's like asking for a unicorn. And so we decided probably the first data executive is going to come in there, make a bunch of changes, get a bunch of people really mad at them, and then take one for the team. And then the second person can come along after they've made all that hard work and do it. That's not a very good message, but we'll see how that works out. We're actually studying that. So there are no unicorns out there. There are some really good people and some really good talents in people. And if you look, you can find them, but you're certainly not going to get it through a typical hiring panel process. Remember, there is never any less data than there is right at the moment. Now, I introduced the data doctrine last time. I'm just going to briefly bring it back up again here to talk about what we're trying to do is to say that there are things we need to value more than things we've valued in the past. And they're on the left-hand side of here, obviously in big type, just the way the agile manifesto did. Data programs, proceed software development, stable data structures, proceed stable code, shared data, preceding completed software, data reuse, preceding completed reusable code. There is value to the items on the right, but the items on the left have to be more valuable. And if you like that, we'd love you to join the dialogue that we're having out there by visiting the data doctrine.com and just adding your name to the list. And if you want to, there's a button you can check and we'll be in touch because this is the first version of this. We never get it perfect on the first version. Somebody's going to come along just like that diagram I showed earlier and give us an improved version of this and that is great and we'd love to do it. We want to make sure that everybody can share in that so that as we get new things, we'll be able to put them out there. So those are the prerequisites that we need to have. Now let's talk about data strategy iterations and these are particularly important. So we are here. We've done the prerequisites. Now here's the lather, rinse, and repeat phase. It sounds a little bit trite, but I need to tell you a story. When my wife and I first got together 12 years ago, she said to me, look, you talk business, she's an accountant. She says, we cannot have a business conversation until you read a certain book. I said, okay. And she said, if you don't read this book, we are having no business conversations period. Okay, tell me what's the book? And the book is a book called And this is the method that we've used here. It's called the theory of constraints. You identify the thing that is primarily blocking you from achieving your strategy. You exploit that constraint in order to make it the most important thing that you're doing. You subordinate everything else to that one constraint and you elevate that data constraint up so that it can then be turned into an advantage. And then you repeat. And I know that's very quick. We're not going to do it that quickly. I'm going to go through the whole process again. And I'm going to do it a third time with an example for you. But the idea works. We've been doing this now for a number of years on how to do it. So here is the theory of constraints process. Again, our data strategy is X. Our organizational strategy is Y. The data strategy should be to support Y with X. And we want to find out what is the thing that is blocking us. It may be that our people are data illiterate. So they're for the first piece of data strategy. Maybe we have to make our people more literate. Or we may have a mainframe computer that is spitting out bad data late. Again, there's a number of different things that can be occurring here. You want to be able to grab whatever it is and utilize it. Once we have exploited that constraint, understood what it is that we're doing and how that constraint is impacting our ability to do strategy, you make everything else subordinate to that. After all, if you're out in the desert and you need water, your first task becomes finding water, right? We're not quite that dire, but most organizations have an immense amount of complexity. Think of it another way, cleaning your house. You don't go in and say, I'm going to clean all the rooms at once. You go in one room by room and clean them one piece at a time. Elevate that constraint, repeat the process, and keep repeating the process as you do, which is how we get to ladder, rinse, and repeat. This process of doing this, I'm going to give you an example here. This case study is available online. You'll get this link after we finish here when Shannon sends out the stuff from the seminar. You can download this entire case study here for free and take a look at it. I'm just going to give you a portion of this. This is from a company called BigCo. BigCo had a really interesting challenge. They had, in their data strategy, excuse me, in their organizational strategy, a strategy of growing by growth. I know that sounds redundant, but we're going to keep growing because that's how we're going to pay for everything there. In other words, a rising tide floats all the boats that are out there. The first thing was they needed to fix the prerequisites. We already talked about what those prerequisites were, the phase one piece. Now we're talking about phase two in here. The problem was very, very straightforward. They had to issue new locations. Their location definition was a store, a store number. They had a very simple process. This was a woman. She is a person who exists at this location, and she had a spreadsheet. That spreadsheet, she took the next number and handed it to whoever came in and said, I need a new store number. That process worked really well. They never had a duplicate store number. It made perfectly good sense. As you can imagine, store number was embedded throughout all of their major IT systems because of course you want to ship to the store number. You want to do payroll by the store number. You want to do all of these other things. The interesting part was they had expanded the store number from two digits to three digits just a couple of years ago. You may ask yourself, why didn't they go to four digits or ten digits after all storage is cheap these days? We don't know. That was before our time, so we don't know the answer to that, but it proved to be a problematic decision because they were running out of store numbers again. If they didn't have a new store number, they couldn't open a new store literally. That sounds crazy, but that is literally what happened. It was a numeric field. They couldn't put alphanumerics in the way the airlines do with your record locator number. There's really more letters than numbers and all that sort of thing. They knew they didn't have 999 stores. They said, what's going on here? What was the data governance problem? The person who was issuing these was told to give out store numbers to anybody who came in and asked for them, but here's what people did. They used store numbers as location numbers. When they wanted mail delivered to a certain spot, they would say give me a store number because they knew they could use that for a mail number. We found store numbers that were for the red boxes outside the stores. The movie boxes where you do retro DVDs and things like that. We found store numbers for warehouses. We found store numbers for all kinds of things that weren't stores, and yet they knew they had not 999 stores, but some number much less than that. On the other hand, they were not able to go back and issue new store numbers because they were getting close to 999 stores. We spent a lot of time in emergency meetings where we would go in and, okay, the red box can't have its own store number. We'd go back and take these things back. Here you can use 145, and here's 213, and we found 646 wasn't being used and all this sort of thing. That just sounds absolutely crazy, but that's what happens when you do things that are spreadsheet-based, that are pervasive across the entire organization, and most importantly, not governed appropriately in order to do this. So the last store number was going to be issued in this calendar year. Oh, my goodness. What are we going to do? Well, the solution from a data perspective, from a strategy perspective, was to say this is the only important IT project that is going on in this company this year until we get this fixed. That made a lot of people pretty mad for starters. There was a brand new supply chain network that was going to go into place, a lot of other things. Nope, sorry, can't do it. We have got to have everybody focus on this one problem because we don't know where this is used and we don't know how pervasive the problem is except for the fact that we don't do it. Then they went to 10 digits. They figured 10 digits, they'd get them to the moon. So they're probably okay on that. You know, there are standards we can look at. I can remember back when I was at the Defense Department and we were convincing them that they had to do Y2K and the Comptroller, the Department of Defense, waited until we were done the meeting and said, are you sure four digits is enough? It was a legit question at the time. Probably we're going to be using the old cobalt systems when the year 9999 rolls around, but that's the kind of thinking that got us in the problem in the first place too, wasn't it? So you do have to take some risk in there and understand that. And then subordinate every other project to it, prioritize this one project, and get the remediation in place. Now the lucky thing was we actually had some documentation lying around from when we had done Y2K in this organization many, many years prior to that. Elevate that, execute it, make it happen up there, and then we went and found the next thing. Hopefully that makes sense to everybody because if we are not managing data with guidance, there is not much point in managing it at all. Remember data is our sole non-deplutable, non-degrading, durable strategic use. Let's look at a couple of other things that can help guide the process of figuring out what your next iteration of data strategy should be. And my co-author and I actually went through a number of things and said, you know, really the book should be called Data Strategies because clearly you can see with this process you're not going to have one data strategy, you're going to do one thing, and then the next thing, and then the next thing. Or your data strategy process may be another way to do it, but nobody thinks about it that way. How should the data be used and in which business processes? So over time you can build up the wonderful CRUD matrices that will allow you to understand how the data is shared among the various users, the divisions, the partners, the geography, whatever it is that you're looking at. What processes and procedures should be allowed to make changes to the data? We don't want just anybody to be able to go in and do that sort of thing. Who manages the approval process and then who ensures compliance around it? These are just some of the things that people have to look at. Let's look at a framework in which we can implement the ideas that we've talked about here so far. Data strategy framework, most organizations start out and say I've got some business needs and these are interesting, and we should just go look at those business needs and move them directly into some sort of solution for the organization. It's just wrong. And the reason it's wrong is because as I said earlier, hand in the keys to the Tesla car, to the 16-year-old that has a driving experience, is probably not going to produce a decent outcome for you. The 16-year-old is in some current state, which is that driving sounds cool, but I don't have a lot of good experience with it. So what should we look at? What is the existing state of the organization? Most organizations are in some state. They have some knowledge of this, but they are not really ready to go in. My favorite example of all this is Master Data Management. We see tons of organizations buying into the say, oh, yes, you need an MDM solution in order to solve your problem. Well, that may be the business need that you have, but if your current state of the organization doesn't allow you the discipline to implement MDM, you will not have a good outcome from that experience. They will not match up when you try to do that. And I've seen many organizations where I'm just walking the halls and I'm hearing, well, I didn't know where to stick the data, so I stuck it in the MDM. At that point, you know, you've lost because that is not the use MDMs are set up for. We can go to the Master Data Management seminar. We're ready for that one to take a look at. Once you have something that actually matches the current state with your business needs, oh, and by the way, in that situation, what I tell organizations is don't go spend tens of millions of dollars buying an MDM package from all the vendors. The packages work well. That's not the issue. The technology works. The organization's ability to put it to use. What happens is I say go out and build yourself your own little MDM-like capabilities using something that you're familiar with. Most organizations have a good deal of SQL server capabilities around. You can build yourself a little and not don't buy MDM stack from Microsoft. I'm talking about just build and Master Data Management solution in this. You will learn so much in a year or two of working with that system that you'll then be able to have a conversation with the vendor about the overall process. It's very critical to balance those business needs with the existing capability of the organization because you can make all the roadmaps that you want, but if you don't have one that actually you can drive on, it's not going to work. The other part of the strategy piece is to make sure that you balance the business value that is derived from your strategy implementation with new capabilities. The organization has to have some balance between tangible business value and also organizational capabilities. Again, in the MDM solution, and I'm really hot on this topic because I've seen it messed up so many times by so many organizations, is that when the MDM consultants leave the organization and the organization didn't end up learning, absorbing from the consultants what they needed to learn, then they kind of go, oh, I don't really know what to do with it, but I can apply patches to it and it'll run there, but I heard a story actually up here that there's one organization that they've put the MDM up, but I think nobody's using it anymore. It's just spinning using the license over and over again. I don't know where I heard that, but it's definitely on this trip so far. So balance that business value. Make sure that your strategy delivers something that the boss can say, oh, I see what you do here and it's nice that you do that for me on a regular basis so I don't have to come ask you, what is it that you do again? On the other hand, we've also got to build the organizational capabilities. How are we going to get people better at doing this if we don't start to bring this up? The next and the last thing for today for this is that the data strategy needs to be sequenced. And when I say sequence, strategy is always about choices and the most basic choice in looking at a strategy is between innovation and improved operations. This chart is shamelessly ripped off from Michael Porter's work. He's done a great job of proving that these are the only two places that you can do this. And what that means is that most organizations start out sort of here with not really much of a strategy. A good place to go is over here to quadrant two, which says let's focus on efficiencies and effectiveness. And of course, Walmart is absolutely king of efficiency and effectiveness. With a strategy of everyday low price, what else would they be good at? On the other hand, some people think Apple is kind of innovative in their mode, so they look at this opportunity to create strategic opportunities by putting it up in that quarter. What doesn't work is trying both of them at once. What works even more poorly than that is to take the people in quadrant three who are really good at innovation and tell them to be effective and efficient. Or to take the people that are down in quadrant two and say let's not be effective and efficient anymore, let's be innovative. It just doesn't work. If you train people to be effective and efficient, they're not going to be able to be innovative with a snap of the fingers. They can get there. The reason I like to sequence it in this fashion is because if you go from one to two, you can take the savings at two and use it to fund the things that are up there, fund at three. So that's what I mean by sequencing data strategy. Then eventually you might get good at this, although I've seen maybe a handful of organizations be good at these in all the years I've been working with them. And only one in 10 organizations is actually any good at this anyway. Change is hard. And many of you have seen this diagram before. It's just simply to remind us that we can do all of the things that we want to in the world, but that we've got to make sure that we hit the cultural aspects. If we don't hit the cultural aspects of this, nothing else that we can do will work in that process. So we're getting back up here towards the top of the hour. Let me just sort of review where we've been. Again, I like these statements. So hopefully you guys can help me improve on them. A data strategy specifies how data assets are used to be used to support the organizational strategy. And we look briefly at strategy. We look at what is a data strategy and how they work together. After you have your data strategy, your data strategy is necessary input for effective data governance. You use your data strategy for three things. You improve your organization's data. You improve the way people use their data. And you improve how people use the data to support your organizational strategy. It is kind of crazy to try and do all of that at once. So again, sequencing makes sense here as we're moving out. It's very important that you understand that most of your organization doesn't know any of this just yet. Not through any fault of their own. Some have learned it over time. John and I were talking about how much time you spent in IT and data sort of kept coming up. It's a pattern we see over and over and over again in here. But the lack of the organizational readiness, the lack of data competencies in the organization, and those seven deadly data sins are really, really problematic all the way around. Once you have eliminated them, then you can start on the lather rinse and repeat cycle. And again, remember that is, once I've got the organization ready to go, once they are data literate, and once I've got the seven deadly data sins out of the way, I can then turn around and say, what's the next thing that is keeping me from doing more with my data to get the organization further down the mission? If I have that question that I can answer very quickly, I can put the right people and the right tasks and the right technologies to work on it and say, you know, let's play with this SQL Server MDM-like solution for a couple of years, and then we'll go talk to the vendors. And you will be in a much better position to have a very good conversation with the vendors at that point. Remember also that it's important to have a balanced approach that you can't have it too much towards value because it will never deliver capabilities and too much towards capabilities because it will never deliver value. You've got to be able to say, the boss has got to be able to say, your boss is always asked, what do those people over in the corner do? If the boss doesn't say, ah, yeah, remember that $20 million problem they solved? I'm sure that's the group that solved that problem. They can't say that. They're going to have a hard time defending you on that. So we're now right at the top of the hour for our Q&A. Let's see what sort of questions you guys have for me for the first time out on this material. Peter, thank you so much, and I love it. Another great presentation. Great way to kick off the year for the series. Just to answer the most commonly asked questions, just to let everybody know, and to remind everyone, I will be sending a follow-up email by End of Days Thursday with links to the slides, links to the recording, and anything else requested throughout the session. And, you know, Peter, we've been getting a lot of industry-specific questions, and here's another one. Any case histories in the mineral processing, engineering, construction, chemical processing, and the environmental industries? Let's throw them all in there just to be safe. Why not? One of the first things I love to direct people to in the mineral industry is the case of Gold Corp. Shannon, do you remember the author that came in to talk to us a couple of years ago that wrote Freakonomics or Wikonomics? Oh, yeah. Freakonomics is Wikonomics. Dan, give me a second. I could probably pull up his chart here. He did a phenomenal case study in that area where they had a gold company that was doing this. All right, give me just a quick second here. So he's got a gold corporation. And when you do gold mining, of course, you want to get all the gold out of the mine. You definitely don't want to leave anything lying around for others to get. And he thought, by his estimates, that his gold mine was, I think he thought it was 90% out and that he didn't need to have it anymore. There wasn't any more gold that he could find in it. Dan Taps got that too. Sorry, I was looking for the name and it finally came to me. So what this fellow did, he actually went to an MIT conference on Linux for some reason. And he said, oh my goodness, I shouldn't actually rely on my own experts to tell me whether I'm out of gold or not. I should put it all on the web. Now, in the gold mining industry, the last thing you ever do is put your mineral data on the web. Right? That's just no. We never do things like that. Well, he did. And he offered cash prizes for people who would have better ways of approaching gold. And they found another 90% gold in this mine that he thought had been all tapped out. Because he had people with physics degree. He actually even had somebody with astrology credentials come in and find some gold for him, which sounds completely crazy from the process. So that's a great story from the industry. I've actually been working in that space for a little bit. What you'll find is that that mineral data, what they're trying to do is to say, how fertile is the soil going to be? What can we find in that soil? Is there going to be oil down there that we can dig on? And they've gotten pretty darn good at this. So there's actually a couple of events coming up, I think. They may be on the university site to take a look at them that are coming up. But anyway, there are definitely some case studies out there, and that's what I would do. Dan Tabscott, I think is his name. Tabscott was the guy that wrote the book. Thank you. And included in that follow-up email, I have a link to all past webinars and stuff for this series specifically. So it should be a little easier to find that way. And could you repeat the author of the goal? The goal. Ellie, who's Goldrod? If you just Google the goal, there are so many used copies of it out there. My students love this book, because you can pick it up for like a dollar. I love it. Are there any case studies with geospatial domain? Are you there? Yes. Once again, when you think about, well, let's just talk about, first of all, the value of geospatial data. Most people aren't aware that IBM was looking to buy weather data, and instead they ended up buying the weather company. And they found it so valuable that there was going to be some tremendous, tremendous opportunities in there, and they went and grabbed the entire company. And so IBM now owns like weather.com or something like that. When you think about what's happening on the planet, being able to forecast it. Again, I'll go back to Walmart. Walmart actually is in some ways better at forecasting what disasters than the federal government. They've looked around and said, you know, if we preposition water bottles in flood areas, A, we can do some good, which is a good thing for them, but we can also make money on it as well and be helpful. So they'd rather sell their supplies than have targets sell their supplies. So Walmart competes with the Weather Service to actually get ahead of storms and get people to look at what's going on out there in terms of storm barriers and things like that. Oh, yeah, someone just posted here the link to the book and you get it for $1.99 for the goal. There you go. Thank you. I think that's all the questions so far. I'll give everybody a couple of minutes. We're only five minutes into the Q&A here. Otherwise I think you've given everyone a lot to think about here, Peter. We can talk about next week's seminar or next month's seminar. Absolutely. I'm excited about the agenda we've got coming up. Constructing your data garden. I wonder who came up with that title for us. That's not what, to me. Yeah, we're really looking forward. This will be a... I wasn't here. Yeah, okay. We did this collaboratively on these things, but yeah, we're looking at constructing your data garden. Let me ask anybody who's got any questions too. Clear as mud, right? Maybe we just overwhelmed everybody, Shannon. I think so. Way to go. Lots of compliments on the presentation and information, and a lot of people asking for the slides. I know you've been heard. I just wanted to know if there's a question. Everybody, I do think that Peter... Try this step out. Yeah. And Peter's information is included in the follow-up email. So if you have additional questions that you think of later, always happy to get to that. So Peter, thank you for kicking off the new year with a great topic and presentation. Loved it. And again, I also had a follow-up email to all registrants by end of day Thursday with links to the slides and the recording. And I hope everyone has a great day. Thanks to all of our attendees for joining us in this brand new year, and we look forward to seeing you next month. Peter, thank you. Thanks, Shannon. Bye, all. Have a good one.