We're gonna be talking about specifically, BI methodologies, the value that they help bring, the goal you should have for them, and how different factors influence how effective you are at achieving those and how they might influence what type of methodology you may have inadvertently ended up with. So there's a lot to cover here, but broadly what we would consider this is we're gonna be talking about strategy and the direction and vision that you're setting for what you want your data and analytics to do for you. And the way that that imprints into your organization in terms of people and processes and platforms is critical for how you think about your methodology. So that's what we're gonna cover today. We are recording this, so if there are colleagues that, were unable to make it or for whatever reason you wanna revisit this content, it'll be on the inter works dot com website in about a week. I'm Robert Curtis. I'm the managing director for Asia Pacific at InterWorks. Just scanning through the list of attendees here. Several of you, I already know. You are friends of ours, so thank you for, joining us. There are some new faces. So if this is the first time, nice to meet you. A little bit about InterWorks. I won't spend too much time. Remember, get two slides just to sort of give you a little bit of who we are. It's important to understand who you're talking to about this type of stuff so that we've got some level of credibility. Basically, we we do strategy solutions and support for data and analytics. So we can help define what your vision of where you wanna go is, and that could be across anything related to data and analytics, whether it's your cloud strategy, your enablement strategy for your team, etcetera. We build solutions, and we we do those collaboratively. We can do a lot of the heavy lifting, or we can do a lot of the mentorship so that you can do it. We're very flexible in that approach. And then also we do support, and that might be support for your tools and platforms. It might be support for the solutions we build or support for your users and communities. We really try to do a full three sixty on everything that you need. Some context on on how successful we've been at that. We've been in business for twenty seven years, which is we have long gray beards when it comes to the consulting world. Most consultants, that start their own firm do this for ten years and then sell to one of the big four. We like what we do, and we wanna continue to do it the way we like doing it. So we've stayed true to our mission and have remained independent for the entirety of our existence. And that's led to a lot of success. Seventy five of the Fortune one hundred are our customers. We are global. And as a result, we have a lot of, visibility. One of those is our data blog, a data and analytics blog of all things generates over three point five million page views every year, and it has for at least a decade. We have somewhere on the neighborhood of eight thousand clients across every industry vertical you can imagine. A lot of those are here in My Remit in Asia Pacific, but we've got friends in Europe and the United States. So we've got a tremendous amount of experience with customers of all sizes. And a couple years ago, Forbes went and picked twenty five companies that they considered small giants across different, you know, whether it's services or whatever, across different whether technology consulting or retail or whatever, and we were one of those. So that was a great honor for us. We work with a lot of great partners to do things for a lot of great clients. This is just a snippet of some of the people that we've gotten to work with. And there's, of course, we are we are we have tools that we really, really like, but more importantly than anything, it's less about the tools and more about helping people. And so that's why we have a lot of different logos on there because there's a lot of different ways we can help. And depending on what your needs and your use case, different tools might be a better fit. So we're pretty open to that. Alright. Enough of the preamble. Let's get started. So I'm gonna kick this off by asking you a a very important question, which should be at the heart of every time you sit and you think about data and analytics. Anytime you're sort of mapping out strategy or vision or your future state, the number one question you should ask yourself is how do we, as an organization, define and determine value when we're thinking about data and analytics? And there's a lot of ways you can do this. And depending on who you ask, your COO, your CFO, whatever, you will get to different answers. But I would I would pose to you that there is a right answer here that then you can apply those metrics to with a lens, and that's user adoption. Broadly speaking, and there's an asterisk here that I'll get back to in just a second, good things will happen when users use stuff. For us, it's when they use our data or they use our data applications or they use AI or they use our analytics or data science or whatever. Now the asterisk is, of course, they have to use it the right way. So that's why I'm gonna say can happen when they use stuff. If it's ungoverned and we just sorta let them use whatever data they want in whatever fashion they want, that's how stuff gets broken or you end up in a headline. But generally speaking, if we give people the proper pathways and support and enablement and then turn them loose on our data and analytics, we're going to get good results. And overwhelmingly, I would say this is what you should be thinking about when you make any investment in technology. So what does, adoption strategy look like? How do you focus on user adoption? Because it really should be an effort, a a specific measurable strategy that you are trying to execute just as much as if you were doing a migration from on prem to AWS or this tool to that tool or we're building a customer success partnership. Whatever it whatever material thing that you're trying to do, think of it very much in a project format as you would with user adoption. And I would say there's a couple things to think about. So again, we wanna bring users into our tools, onto our platform with the following considerations. The appropriate BI methodology. Different BI methodologies are suitable for different organizations, different maturity of your data, as well as different goals that you have. And there's no right answer. There are answers that are less right for you or more right for you than there is this is the end all be all for everybody. A complementing tool set. We will do a little bit of comparison of some of the more prominent BI tools and how well they fit a particular methodology, or if released into the wild, what sort of methodology they naturally start to develop. Well crafted governance policies. This is the heart of anytime you have users interacting with systems and data. We want to make sure they're using the right data in the right ways. And then comprehensive change management. If we are doing something new or we wanna reinforce good behaviors, we have to make sure we have a well thought out communication strategy to the different types of cohorts that this intersects. So there's a lot there. We can't cover all of this in one session, so we're really gonna focus on the first two today. If there are questions about anything that we cover today or you wanna talk more about governance or change management, we we do those things, and we'd be more than happy to talk about how you're going assessing, as well as planning for future success. So when we're thinking about BI methodologies, the question then is probably naturally, well, how do I know which one I should choose? And I've got this little logo here, this little image here with a little chicken and an egg. This is the idea of which one which one comes first. And I would pose to you that while most of you might ask, how do we choose a BI methodology? The reality is is that probably the methodology chose you. And I see this from twenty plus years of being in this business. We will go and talk to customers, and we'll see the landscape. And immediately, I will know what their problems are when I when I get an understanding of how the user community interacts with data and the people that are being the stewards of that data. It will immediately tell me, okay. You're doing it this way. You've got problems x y z. Or you're doing it that way. You've got problems a b c. And I'll explain this in a second in terms of what those actual methodologies do well and what they don't. But the reality is is that most organizations don't think about BI methodology. They don't pick, plan, or prepare for it. In fact, it happens by accident for most folks. So what do I mean by that? Well, drawing from my own personal experience, I'm gonna give you some examples of things that I get told by customers, by our clients, and I'll tell you, based off of the statements they make, who actually decided on the BI methodology. And very rarely is it a council of well informed stakeholders and SMEs together deciding this is the best direction for us? Let's invest in processes and tools that support this. Most often, it's something like this. IT chose the analytics tool, meaning they've got some buying power for all the other stuff we do, our our hyperscale or our ERP or whatever, and they got the tool as a result of that. This does not mean that IT actually chose your BI methodology. It means the tool did. And I'll explain when we get into the specific ones how some tools naturally take on a methodology that is specific to the way that tool was designed. Here's another one. We'll let the business units or groups of people decide what they wanna do. We don't wanna prescribe anything, which means no one is deciding. It's going to be everyone's going to go in different directions or at least by group, they're all going to go in different directions. We'll cover more of of what this means later too. The next three are all emblematic of some of sort of a similar thing. Peggy, some random person that introduced this tool and everyone liked it. So now everyone is using it or a lot of people are using it. Or no one knows what's happening. We don't know what IT is doing. They've got some big operational reports that everyone can use, but I have things I need, and I don't know who to talk to. So what do I do? Or, you know, they told us something is coming, or I can get this individual report, but I never see it. I keep getting bumped from these other projects. So I'm just gonna go figure it out on my own. Those are all three statements that sort of lead to one to one of three potential particular or maybe all of these, outcomes. One is the single evangelist. Any one of these three could be somebody like, I'm just gonna go find something because I really wanna get this answer. I wanna work with data. I wanna have more insight. I wanna do my job better. So they'll go find something and bring it in. And if they get success and other people are looking over their shoulder, it can become, emblematic of how you're gonna do BI by accident, or your vendor might decide. What do I mean by that? What I what I mean is is that when salespeople talk to your organization, they have strategies for how they would like to introduce the tool. And they know that if they do this way or that way or this way, they have a better chance of getting a sale, and they're incentivized to do this. That's all they think about all the time. And there are many different approaches. There could be the top down approach, which is we're gonna get in with the executives, we're gonna take them golfing, and then they'll buy it for everybody. The good news is is that they have one big sale, one big purchase, and we can enterprise make a decision. The bad news is is that very rarely means that the stakeholders are actually being informed. The other way is land and expand, which is exactly what we see here. Peggy or somebody else is talking with a sales rep from a BI tool. And, like, yeah. Yeah. We'll give you free credits. We'll give you a free application. We'll give you a free license. Just go and start doing stuff. And then once those dashboards get created and enough people using it, it's really hard to get that tool out. And sometimes that leads to great successes. Like, wow. I'm sure I'm glad Peggy was here. Sometimes it cannot. Man, I wish Peggy had actually waited for us to get our solution in place. And the last one is it's a random Google search. How do I do this thing? Whatever ends up coming on top, that's the tool or the way we end up doing it, which kinda leads me to the last one. We've been doing it for so long this way that we've got so much tech debt that we kinda have to do this because it would take too much in time and energy to undo it and do it the way we probably should have done it or the way we would do it today, which means past mistakes, past decisions, and sort of deciding your BI methodology for you. Overwhelmingly, this is what I see when I talk to customers. Very rarely do I see we made an informed decision based off of our goals, future state, and and a consensus of our stakeholders, which is how it should be. Before we go any further, I do wanna talk about measurement, because I think this is important in terms of when we think about success with BI, how do we actually put that into to numbers? So there are a lot of ways to do this, and I'm gonna start with the ones I think are best and then kinda work to the ones that are easier to measure but probably less indicative. But that way, you've got a a a gamut of things that you might think about in terms of how you might measure your success. So the easiest one and the best one I should say the best. The easiest one. The best one, but the hardest one is I did something. It was a project. We completed it and pointing at dollars. We saved this much time, which which can be converted into dollars, or we save this much money by making this efficiency, this automation, whatever. That's great. Executives love it when you can tie projects back to dollars or time. Or the opposite of that is we created value, new things, and this has led to this uplift, this new amount of money coming in. Executives love that, but those are harder. Not every project is going to have a very tangible, we did this, other than there's other ways you might measure that. If you can do those things, great. Dashboard usage metrics. We introduced the tool, and no one's using it. Or half the people are accessing a dashboard once for every ninety days. Or maybe we've got eighty percent of the organization using these dashboards. On on average, they're accessing five different dashboards. Those are indications of how well your tool is being used. You may not be able to point to dollars, but you can justify the expense when you've got a lot of people using your data. There will be other downstream things that come as a result of that. I've got this one. I'll put an asterisk on this. So another way to see how well your data is being used is to if you've got a consumption based or credit based platform like a Snowflake or a Databricks, you can see how much people are consuming on the platform side of this, on the data side of this. With the asterisk, people could be assuming a lot because maybe they haven't optimized their their compute, and maybe they're running these queries at a much larger compute capacity they need or much longer than they need. But assuming all of those things are tightly database or your data data cloud platform is an indication that there is usage. Logins in the last ninety days, this is easier to quantify. Community of practice events and attendance, that is a good sign of data culture, and data culture is user adoption. Data culture is an asset. And then other things like subscriptions to platform output. So for instance, somebody made me a dashboard. I have a subscription that shows me that dashboard results every morning. In aggregate, that gives you a good sense of people are using this information. I've I've built a tool to get people more data, to make data informed decisions, and so there's a variety of ways I might think about that. So So hopefully, this gives you some starting points if you're starting to think, how do we actually measure our payoff for investing in all this tech? Here's a gamut, a scale, if it helps. I would say that when you are thinking about choosing a BI methodology, I have one overriding guiding principle. The the thing that you should be thinking about as the preeminent, goal achievement future state, and that's self-service. Your goal should be to convert as many business users, and I'll define what those things mean in a second, into data workers as you can. Most tools, depending on the tool, achieve twenty to thirty and a stretch fifty percent usage. Meaning, let's say that's they log in every thirty days to your platform. That might not even be creating dashboards or the ability to. But if we could get a hundred percent let's be let's be crazy. If we get a hundred percent of your users with the ability to ask their own questions of your data in a way that gives them predictably accurate, results that then influences their insight into their particular aspect of their business and drives behaviors, that is a very, very powerful thing. I can't say this enough. The transform the transformational power of a data culture cannot be overstated. It is critically valuable. It is very difficult to do, and it it is not something that we do and then we're done. It is something that it is like garden that must be watered and weeded and and curated. And a lot of people agree with my point of view. Here's an example. In a world of more data, the companies with more data literate people are the ones that are going to win, and this only becomes true, ever more true with AI. The more people that understand what's happening and how to use it to their advantage versus waiting become the people that are then going to revolutionize the way you think about your customers, the way you think about your business, the way you think about your revenue and your costs. So why don't more people do this? Why why isn't self-service more attainable or the degree of success to attain sixty percent of the folks? Why does it happen more often? Well, this comes into the idea of these BI methodologies, and there's a lot of different ways you could think about these. There's there's probably a dozen niche BI methodologies. We're gonna cover four four big ones, And I tried to do them in a way that they might mature into one another. Now there might be some that skip steps. We'll talk about that on a subsequent slide. But the way that you organize how people interact with your SMEs, how your users interact with your data workers, and then how they interact with your platforms for analytics and those interact with data are all very important. It will either give you starting advantage for for some things that you're trying to accomplish or it might set you back on some other things. So you have to weigh the pros and cons of each of these methodologies. Before we get specifically into these, we're gonna do some little definitions here. So when I say business user, I really mean somebody that has dependent data usage, meaning they need somebody else to help them answer their data questions. I use the term business user. If you're a not for profit or you're a government person or you're in IT, but you don't interact with data, I'm referring to you as a business user. You'll be the gray icon. For those people that are empowered to ask data questions because they've got access and technology, they're purple. We'll call them data workers. And then as a result of the various things that have to be done, there are analytics assets that must be created, reports, dashboards, whatever. The business user is dependent on other people creating them for them. The data worker is not. And then there's also data assets, whether that's your business logic or your data pipelines or whatever. Those are things that have to be built and maintained as well. So this is how we're gonna evaluate this in the context of how close to self service can we get, we want as many of these gray users to be purple as possible. That's really the goal in a way that's governed and predictable, etcetera. Okay. So let's start off. And by the way, if there's any time you have questions, I think there's a chat or maybe there's a, q and a in there. Chuck it in there. If I happen to see it while I'm presenting, I might stop and answer it, but I'll try to give us about ten minutes or so. It's five at a stretch, if we go long, to try to come back and answer. And, again, if you have any questions that I can't get to, we have, a whole bunch of account executives and solutions folks that are happy to talk to you. So the first one, decentralized BI or what people call the wild west. And when people think about self-service, a lot of times they think about this. Self-service does not mean good luck. That's not what it means, but most people start here. And so this isn't really a methodology as much as it is the absence of one, but I think it's important to talk about the challenges that you have if you don't put thought into this. So there's no standardized approach to analytics or data and no centralized support for the end users across the enterprise. There may be reports that are created for everyone to use without customization or consideration of what their individual needs are. And a lot of times, those skew quite heavily towards your executives. So I might be a frontline manager. I might not have the ability to get my answers, which also means there's probably very little governance among the user community, and there's certainly no strategy. But this is where most people start. Enterprise data, in terms of how it interacts with self-service, is an untapped asset. And because we haven't made big investment, we're not getting the economies of scale for the money you could invest in the tools, getting the most bang for your buck, or investing in that data culture to where there's this sort of rising tide of more and more people understanding data and our data literate and understand how to do that with the tools you've invested. Business logic is often, well, Peggy knows how to do that, and Peggy's on PTO, so we're crippled until she comes back. Or Peggy left our business, and, oh my gosh, she walked out the door with all of this understanding of how to do stuff because we didn't have processes and governance and documentation. And when you think about driving innovation with data and analytics, it's often individual evangelists that are doing this, which is uncontrollable and unpredictable. So we've got our little diagram here. Let's go into how this actually manifests. We've got some sample business units, and then we've got sort of IT. And so what this looks like is IT is largely focused on operational reporting for everybody, not the individual, not self-service, supporting platforms and applications. These are the folks that look after your your ERPs, your CRMs, your host, your on prem servers. A lot of times, they are completely divorced from the business. Their interaction because it's the users because it's the users that are driving the advancement. How easy can these tools be for the rest of the organization to leverage them and solve exciting new use cases, which which really means overall from an IT perspective, from an organizational perspective, there's a lot of lost opportunity here. Your users are not driving value for you. And I've got the purple user sort of sprinkled randomly through those business units. And again, these are the people that are not settling. They're trying to solve these problems, and they should be rewarded. I did a, a couple presentations where I sort of presented some armchair organizational theory, and there were four states. There was the lone wolf, the wolf pack, the herd of cats, and the swarm of mice. This is where you get your lone wolves. The people that are like, I wanna go do something. And they end up going and doing great things, but often without any coordination. And so for them, the tool of choice is random. Enterprising evangelists may appear randomly throughout the enterprise. Use cases are often divorced from our technical capabilities, meaning I wanna solve a thing. I don't have the ability to do it. I don't have any help, so I'm gonna do the best I can with the tools that I know how to use. And sometimes these evangelists in b u one and two and three may find each other and start to build a little bit momentum, but oftentimes, they may not. And the overwhelmingly, the tool of choice here is Excel. It is a BI tool. This is not a very good one. The other tool that you commonly see here is when somebody buys a Microsoft license, everybody gets free Power BI on their desktop. And so for those folks that are really ambitious, they might start to do their own localized desktop Power BI. It's harder to do, because the technical requirements to use Power BI effectively are are high. So most people do analysis in Excel. Again, completely ungoverned, completely invisible to the rest of the org. And when you're thinking about data breaches, this is front and center where these things happen because nobody knows what Peggy I keep picking on Peggy. Larry we'll pick Larry. Larry had in his Excel sheet that he sent to some vendor. And, oh my gosh, he didn't think about the twenty five worksheets he had in that workbook. You can see how these things compound. So that's your first sort of starting block state, and a lot of organizations sit in this particularly more conservative organizations that don't have a need to be data focused, sit here for a long time. What naturally sort of evolves from that is eventually someone important says, you know, we should probably just get IT to take care of this. It's data. They're data people. Let's put everything into IT. It doesn't necessarily have to evolve from there in that way, but often that's what you see, a very centralizing where we we're we're reacting exactly the opposite. It was very uncontrolled. It was wild, wild west. Let's centralize it. And that that that that there are some good things here. One, it's very highly governed. Nothing gets out that IT doesn't get to look at, which means they're gonna use the best data. They're probably going to build very, very accurate, dashboards. The challenge though is a couple. One, those dashboards are probably not very specific to the business users because IT isn't across. They're not the experts on that area of the business. They're the experts on the IT tools, which means there's a division between how IT understands the use case versus what the business people understand. Also, as we'll see a bit more, this is a very slow way to innovate and to iterate. And because IT is so task heavy, they have a lot of things to do. Let's take a look and see how this actually plays out. So here we've got all of our data workers in IT. It doesn't necessarily have to be quote, unquote IT. It could be a dedicated analytics unit that is, again, separate from everyone else that's serving as your data workers. Whether this is a centralized team or you have a consultant approach, meaning you pay them internally as a cost center to then do work on your project, it doesn't matter. It's a centralized analytics team that then has to manage the queue of of requests. And so we talk about queues. Let's say each one of these business users down here has a data need. So each one of them has some sort of dashboard report that they need specific to their business unit or some variation or customization. And let's say roughly every three dashboards has a data source. These are completely made up numbers. Might be more. Might be less. But IT is the one that's shouldering the entirety of that queue. So IT is often overutilized. I should say ITBI is often overutilized. They have a lot of tasks. Build me this dashboard. Oh, no. Executives want something. Everything gets put on hold, but we go fix this thing. Oh, no. We did we have a support function as well. We have to stop all of our BI stuff to go do this thing. A lot of times when you see this, there's not even three people that can do this. I've seen plenty serving large organizations with a single person that is Tableau server administrator, Tableau dashboard builder, as well as the data engineer. And as a result of this being centralized with IT, often IT is the one that defines the strategy. And you could imagine their strategy is gonna be, let's make it it simpler because we have too much to do. Is that really in the best needs of the business where you have one perspective of a team that's shouldering all of the tasks versus understanding the value. And because all of these people down here in the business units are the ones with the understanding of the business, and they don't have the technical means to go and ideate. There is another division there, which means it's very slow to advance and get smart and sophisticated with the questions you're asking of your data because you are segregating them. And down here for your users, they're often frustrated. Long queues means zero iteration. I just got this dashboard that I requested a month ago, and I know if I go and ask for some changes, but they didn't quite get it right, I'm gonna have to wait another month. And, again, the use cases are divorced from the capabilities, which means we are we are crawling versus walking or running into our analytics future. Hopefully, as I go through these, there's people like that's us or that was us, and, oh my gosh, I'm glad we're not there anymore. You often see this when you release Power BI, even even sort of the the, the more centralized SaaS version of Power BI. And the reason that is and I'm I'm I'm talking historically, is because without a lot of extra effort, Power BI is a difficult tool to master primarily because of DAX. DAX is a language that IT people are comfortable with, not business users. It's too far away from Excel and too close to a coding language for business users to get really dangerous with this on average. You are undoubtedly exceptions to this, and some of you might be business users like I love Power BI. But when you think about this in aggregate as a trend, when you release Power BI without a plan, you end up with this. And I will talk to a lot of organizations that will say we really are committed to self-service. We chose Power BI, but maybe they didn't have the change management plan. Maybe they didn't have an enablement strategy. And two years later, they're still here because it takes exceptional effort to escape the natural trends that Power BI is going to put on you. I'll put an asterisk on that. We're gonna come back to this later because there's things that tools are doing and Power BI is one of those big ones that hopefully changes that momentum. Departmental BI. So a lot of times you'll go from centralized to, hey, Finance, they really, really want to get access to data. Or maybe it's the product team or the customer insights team or whatever. But these guys are really, really motivated, and, man, they are busting down our door for more and more and more. Just let them have their analytics people, or maybe it's two of them. And then centralized IT will continue to support everybody else. This is what I call the haves and the have nots. So the loudest people or the people that have the biggest mandate, they get analytics, but everyone else is supported from a centralized team. And you can kinda see we're in a transitional phase here, and there's some things that are really good about this, particularly for that one department. Let's call them finance. And for the rest of the organization, less so. But, again, not saying this is right, not saying this is wrong. It depends on what you're trying to accomplish. So what this means is you have IT, the center, really leading strategy, but very quickly these can become divergent, And you probably have the situation where you don't just have Power BI. You have Power BI and Qlik or Power BI and Cognos or Power BI and Yellowfin, whatever it might be, because this other team is like that tool best solves our problems over here, and our problems are a little bit different than yours. The challenge is really the challenge is when they start to develop their own data strategy, and they're like, we need this data platform, and you guys can have that one. Do everything you can to resist that that momentum because there will be a point where you're like, you know what? We gotta bring this all together. And, oh my gosh, we'd have to do a full data migration with different metrics, different logic, different workflows, a different ETL tool, and get it into one warehouse so that we can all benefit again. The queue for the select department is faster. They are separate from the business, which means they can act fast and elite and agile, but they also suffer from that separation. Let's take a look what this means in reality. So we've got our business unit. Let's say finance over there, and you can see they've got a data worker there to empower the rest of the folks to iterate faster for them because they've got mandate. They've got priority. There's still a queue for the IT folks. It's smaller because they're servicing less people, but they still have all the enterprise data sources that these guys are probably going to need. Maybe it's their primary data source or maybe it's how they enrich it. At the at the very worst, there's a data icon over here underneath this queue, which means they've got their own data, which means they've got their own business logic, and very quickly it will diverge in ways that are not sustainable. So IT is still overutilized. They're still doing too much. They're trying to do dashboarding and manage data at the same time, heavily task focused. And because there's this rapidly accelerating team that has smaller amounts of work to do to do a full revolution versus the IT, a lot of times the other smaller department is faster in developing strategy. So IT either has to hold the ground where it will be the foundational strategy, or we're just gonna let those guys define it because they're so far ahead of us because they can go faster because they have less to do, which then also means when you think about it, it's still difficult for ideation innovation because these groups down here really have no voice whatsoever. These guys have a loud voice, and IT is not reactive. So your dependent users, they are still frustrated because there are queues. But oftentimes, they are even more frustrated because they're looking at these folks over here, and they're like, well, finance gets to have reports. Us in marketing, we only get the stuff that we get out of the the applications that we're using to track campaigns. It would be great if we could do all this other stuff. And, again, the use cases and the queues mean that you are stagnating your ability to quickly innovate on what you could do. There are a lot of organizations like this, and if your priority is to get rapid acceleration and within a particular group, department, or team, go for it. But there are long term, downsides as well. Let's jump to the next slide. You will often see either Tableau or Power BI here or both. A lot of times, there will be a tool for the enterprise, and these folks over here, because they're rapidly accelerating, will go find their own tool. And, again, some departments like different tools better. You can kinda see finance people lean into a Power BI a little bit more, Whereas Tableau is more for your your marketing and creative folks. I'm not saying that Power BI people are uncreative. I'm saying that what you can do in terms of creating visualization is more limited in Power BI, but it suits people that like to think, from a data preparation standpoint, far more robust on that front. Whereas Tableau, you're not limited to choose these fields, choose this, visualization type. You can go and create a lot of fancy exciting stuff. So distributed. And I put this as self-service with a question mark because this is where a lot of people get to and think that this is self-service. And it is former self servicing, but it is not true self-service yet. There are a lot of good things that happen here, but there's still challenges. Naturally, it's this type of setup where you've got a central team supported by embedded assets. So you could call this embedded BI. There's also a way that you could think of this as federated. If that were the case, there'd probably be arrows back and forth and a less of a dependency on the center if you were to say purely federated. But everybody now has the capability of doing it at the department level, which means that it's very easy for us to say, okay. Let's start to build on consensus and compromise. And these are where your most powerful COEs, your center of excellence has come from. Enterprise data, because a lot of the departments are doing their own heavy lifting for analytics, enterprise data can be an IT or centralized function, which is great. The less we have them building dashboards and the more we have them building curated data sources for everyone to benefit from, this is really good. You also get a great economy of scale in terms of your purchasing power across tools as well as your investment in teaching people how to do this. But for certain tools, this can be quite expensive if you're gonna get everybody an individual license. And, of course, we're talking about Tableau. If you buy an individual viewer for a hundred thousand employees, that adds up really quickly. And this is one of the challenges that they have versus Power BI. So let's go and look at what this looks like in practice. So we've got our centralized IT team or analytics team, and they have data workers. And then every business unit has an embedded or more, data worker. And so, hopefully, the IT folks can focus on the data. We're gonna build great data pipelines. We're gonna make sure the quality is great. We're gonna do all of these data tasks that we were distracted by building other people's analytics before. And this is your best chance to get data to everybody that is useful. So check. But you'll notice down here oops. I've got some more queues down there. There should be more queues. I think that they're going to come. I'm gonna click on my button. There they are. My little animation messed up. Down here, there are still queues, which which means these people down here are still waiting on time and access to their embedded resources in their unit. So this is why I'm like, is this self-service, or is this just closer to it? Which is why it's distributed versus genuine self-service. There are still queues. These people still have to wait on this person. That person understands their business better because he's one of them, but he still is a limited a limited resource. So some of the other things that we we talked about on the previous slide, there's a partnership and strategy because all of these voices together are gonna have a very, very commanding, opinion, which IT then can become a business partner with the rest of the organization. The way that IT helps manage, governance is through the access to the data, but you can really quickly still get a wild, wild west down here. If this business user that is a data worker, this embedded data worker, doesn't have a mind of governance or is it upskilled or empowered to control and think about, We have a thousand dashboards, guys. We use seven of them. What are we doing? The other thing that you have to think about is these tools require tool these these business units to accomplish distributed BI require tools that are easier for non IT users to master, which is why Tableau is where you often see these types of things really take root. Because IT or rather, let's say Tableau was designed for a couple of things. One, dynamic in terms of the visualizations you can build with a low barrier to entry click drag. DAX is not there. It's far more of a function based coding language like Excel, which is far more accessible to business users, which is why this whole data for the people, which was Tableau sort of main thing, really pushed it ahead for about a decade until Power BI started building some more momentum. Now this is not self-service because there are still queues, but there is far more innovation because these teams are now integrated into the data conversation, and they have a better understanding what their tools are capable of. We're just leaving these people a little bit behind. These these these nontechnical, business users. That doesn't mean they don't have the capability. It might mean that their role doesn't require them to have it, or it might mean they don't have the time to do it. And And a lot of times your managers will fall into this this category. We certainly want them to be empowered by data, but do they have the time to sit through three or four days of training and then refresh those skills? Probably not. So when we think about these different things, you can kinda see how you might evolve from decentralized to centralized to departmental to distributed. Now people can skip these steps. They can go from decentralized straight to distributed, or they can go backwards, one or the other. And it might be that it's far more complex when you think about a hybrid approach. But But I wanted to give you the basics of these things and and what kind of problems that we commonly see when people have particular ways they're thinking about how people and data interact with each other. So my question then is, is genuine true self-service where everybody in the org can ask and interrogate data to find solutions and insights? The answer is yes, and that's where we get to AI. AI, I'm sure everyone knows about it. I've been every time I go to any sort of conference, most of the time, people wanna talk about AI, what it means for data or analytics or people or customers, is what what what does it mean? What can it do for us? What are the things I need to be worried about? What's the value that it can bring? Well, obviously, AI can bring solutions. Right? It can bring, you can build chat bots for your customers or whatever, but we're gonna talk specifically in the context of these BI methodologies. And so one thing that that AI, generative AI in particular can do through natural language is it can help business users, not data workers, build reports and dashboards. And you see this with, like, Copilot. There's other tools we'll talk about that are specifically a little bit ahead as they introduce a lot of these things into general release. But Copilot and the stuff that Power BI and Tableau is probably behind them in all of this. I'm not making friends with any of my vendors by being as candid as I am. But everybody is trying to do this race to where they can empower customers with no technical background to type in, show me my profitability by business unit for last year. Boom. Dashboard. And the code and all the calculations. That's something AI can do for you. Process and governance. There's a lot of work, and this is one of the reasons why there's a lot of analytics or I should say there's a lot of, governance tools. One of the ones that we really, really like, is Informatica. They've got a lot of AI built in to help automate the workflow of data to check and scan data quality, to build metadata so that there's contextualization and observability on your data. What that does when AI starts to help that particular processes, now governance and cataloging is something that can predictably be successful. I guarantee you if I were to poll you guys, maybe even half of you would say, yeah. We started a governance project. We can't tell you if it was successful. We can't measure it, or we didn't even finish it because we had that we had a counting grains of sand approach. Alright. This field, what do we wanna say about this field or this data source? What do we wanna say? Very manual approach. AI is removing that. That's very exciting and critical. We did a survey of all of our different customers, and all the people that we see across, that come across us at these marketing conferences, whether it's the data and analytics conference or the snowflake event or whatever it might be. So what are your data challenges? We had over three hundred responses and overwhelmingly, it was data governance. I wrote a white paper. We'll share a link, into the chat when we chat about that. But that is a number one problem by some margin that, data professionals are thinking about. What's in my data that is going to worry me? What what value am I not using in my data? And the last thing here is the semantic layer. This is the next wave that AI is going to help solve for us. Obviously, it will not have human intuition. We need human beings understanding why this is important and why this is not versus this is what I can see historically, so I'm gonna predict what I think is important. But it is coming for the semantic layer, which means there can be a data warehouse. And when all of this comes to maturity, a business user can say, I'd like to prepare this data source to then give me these analytics, and AI will help them do that. Again, it still needs machine learning. It still needs all of these monitoring and retraining all that kind of stuff. But this is the power that AI is going to bring within the context of a BI methodology. There's a lot of other stuff it's going to do, but it's very exciting for folks that are in this industry. So let's go back to our distributed BI methodology, which look like this, and let's make it AI augmented. So we still have the same structure, but now we're gonna add AI into the centralized team, which then allows them to reduce the number of tasks they've got here on this queue, data quality, cataloging, observability, and, again, maybe even the preparation of a semantic layer. From an enterprise perspective, AI should belong to IT, just like the data belongs to IT. But it should be a partnership with the rest of the business on how we're going to use it, whether it's for our customers externally or internally to make our users more empowered. And, again, like I said, it's augmented governance. But we're also gonna add AI down here, and I put it specifically down to those business users because these guys up here, these embedded data workers have the ability to ask questions of the data and get answers. The AI that we're going to unveil to everybody is gonna benefit them from a time efficiency standpoint and maybe even a capability standpoint. But, really, what AI is gonna do is it's gonna empower everybody down there. So it removes the skill gap for your business users. It also means effectively there won't be queues because they can go and ask AI to do it for them, and AI is gonna be far more efficient and more more present to answer those tasks. Oftentimes, there is a hybrid approach. So there are there might be tools that are better serving different things, and a hybrid approach is probably the way the industry is going to lean into until Power BI, and, you know, and releases general intelligence. Hi, Power BI. Hi, Robert. Get me a coffee and then do my analysis for me. I'm joking, of course. But there will be a winner that that that takes a big step forward. It might not be for years yet. So, like, a a hybrid approach across tools by cohort and by needs is probably what you're going to see if you really wanna get all of your users empowered. And when you get down to the bottom users, the And so it becomes more about their ability to do a Google type search into the database to get a specific answer for a point in time question. They could always obviously look for dashboarding, but more likely their appetites are gonna be satiated through search or natural language queries, which means there's no there's no more queues, and it means we've made everybody truly into a knowledge worker. AI has the ability to do that. So when we think about this from a tool perspective, Tableau or Power BI could accomplish most of this. And with what they've got in the road map, they certainly are planning to try to do it. I would say a lot of this is still aspirational with exciting releases and road map. I don't think either one of them has completely knocked it out of the park. And, again, there's a lot to do, so it's a very exciting time. This is not a criticism of anybody. But if you wanted to specifically target that that bottom rung, the least technical but biggest set of your users, you might think about another tool. And there's a lot of tools that do this. One of the tools that we partner with is ThoughtSpot. And so you might have the ability to to give them the ability to just go and type questions into their data, and then boom, they get back an answer or they get back a live board. There are other tools that certainly do that, but ThoughtSpot cut their teeth specifically on that. So right now, on that function, they are ahead of everyone else. But, again, you wait two years, the landscape will be different. ThoughtSpot might be even better. Tableau might have a massive release with Tableau polls. There might be some massive maturity there. It is still very much a there are big pieces that are still gonna fall. It's a very exciting time. It's a time for tool innovation as much it is for your BI methodology. So what to do from here? There's a couple of things that we can do to assist. There are a lot of ways that we could help focus this conversation for you. And the the best version of this is what a what we call an SVR, strategy, vision, or road map. Several of you on this call have done these with us. Some of you are thinking about doing them with us. But, basically, it's two to three weeks of our domain SMEs, me being one of them, a data architect, a platform architect, an enablement specialist, strategy specialist, sit down with your team, and it could be your executives as well as your stakeholders. And we do somewhere around eight to twelve whiteboarding sessions on specific topics. Let's talk about your data and your goals. Let's talk about your platform, your people, your processes, your vision. There might be new topics that you wanna talk about. Let's okay. Let's talk about monetization. Let's talk about, AI and and your readiness for it. We can customize these workshops. But, basically, we then build a plan based off of your bespoke needs and our best practices to then point the way. The next thing is the analytics training needs, analysis and plan, training needs analysis and plan, which means what what kind of methodology you're going for. Let's define your cohorts. Let's figure out what your goals for them are, the whole goals and roles idea, and then let's build what what their gaps are. And this is very clarifying for a lot of organizations. So, yeah, actually, we did really well with our data engineers, but our data scientists need help for them to do the things that we need them to do or some other combination of cohorts. And lastly, we believe that you should have flexibility in in in what tools you're using, and that might be singular tool and might be plural tool. This tool is great for finance. This tool is great for our broad users. So if you are thinking about having flexibility and portability in your data across your different tools, we can help explore what the impact and effort might be to do that. If you do have multiple tools, one thing that's not on here is we have a platform called Curator, which is a portal that can sit across multiple different types of analytics tools, Power BI, Tableau, ThoughtSpot, etcetera, etcetera, and integrate them into a single platform. So they can all be in one place versus you having to remember, oh my gosh. Was that a Power BI report? Was it a Tableau report? Was it a Yellowfin report? Whatever it might be, we can get everything into a single place in addition to your non analytics content, that is for data workers. It could be your data dictionaries. It could be your training materials. It could be your community events, all kinds of stuff. Think of it as a content management platform for analytics. Very powerful. That's called Curator. So if any of these things interest you, please reach out. You can scan that code and schedule some time to talk to one of our folks. We've got people all over Australia. We've got people in Southeast Asia. And if you happen to be on this conversation and you're from Europe or America, we're global. We can help you anywhere. I hope you found this useful. If you've got ideas for other presentations, we'd certainly love to hear it. Thank you very much. Have a great day.