 Hello, thank you for participating in our data governance education series. My name is Kelly O'Neill and I'm the founder and CEO of First New Francisco Partners. We are an information management consulting company. Today we're going to talk about some very fundamental concepts around data governance, such as the data governance framework, and a simple process to get started governing data. Let me also go through what the future modules in this education series will be. Module number two is introducing the idea of an operating model, which is the fundamental structure for creating a data governance organization. Module three then drills down deeper into the roles and responsibilities of the participants of the data governance program. And then module four goes even deeper into the role of one of the critical players, the data steward. And module five will go through the process of how to create a data governance policy and tie it into the guiding principles of the program. Next, we'll discuss general best practices for a successful, sustainable data governance program. And the last module, module seven wraps up the series by discussing data governance and its relationship to other data management activities like metadata management and master data management. And we hope you join us for the whole series. So the purpose today is the starting point for data governance and going through some of the basics. Hopefully by the end of the program you'll understand the critical success factors for establishing a data governance discipline within your organization and that you'll understand the key reasons to implement data governance, why it's critical, why it's important. And at the same time, we'll talk about some things to avoid and some things not to do. So because data governance has become a more understood and sophisticated discipline in the industry, we do have several learnings from previous experience that we can share with you and other folks in order to help you not learn things the hard way, like we have done over many, many years. So the process that we'll go through in this section is that we will first start with some very fundamental concepts around data governance. We'll talk about why it's important to definition of governance and then move into some data governance frameworks to help establish the context of data governance. We'll look at the reasons that organizations establish data governance. So what are the business drivers? And then as I mentioned, we'll go through what are some common risks and pitfalls. And then of course, what are some critical success factors for data governance? We'll wrap up this module by talking about how to get started. So back to the basics, getting started with your data governance program. Okay, fundamentals of data governance. Why is data governance important? Well, as I was establishing this presentation and writing down the speaker notes, my five-year-old daughter was sitting next to me and reading through my slides. And funnily enough, she says, data. Data, data, data, data. Mommy, data, data. And I said, you know what, honey? You are absolutely right. It is data, data, data. And I think that that is really very telling. So out of the mouths of babes, right? So really, the world is becoming much more data-intensive. And our world has changed and is continuing to change. So companies are trying to understand their customer and their interactions with their customer at any time, across any channel. Data quality issues that are created across these multiple channels tend to perpetuate even further within an organization. So a misspelling of a customer name, excuse me, a customer name becomes persistent across all systems, potentially in a data governance, without a data governance guidelines and policies. Technology is changing. So many organizations still have legacy mainframe systems that they're trying to integrate with new technologies and even new approaches to managing systems and technologies like the cloud. So as companies move towards a cloud environment, this reduces the level of control around your data and also around the people and processes that are responsible for managing the data. Of course, big data is the latest trend in our industry and this is all a reaction to the fact that data volumes, velocity and variety are increasing. And that data is being created and consumed by mobile applications and in a mobile environment. Project management offices are becoming standard ways to implement new systems, products and processes. And on top of all of this, there's external pressures in the form of new regulations such as Basel III and Dodd-Frank in the financial services industry, the Physicians Payments and Sunshine Act, and the Unique Device Identification in the health and life sciences industry. So these regulations are compounding the internal pressures that organizations are facing in regards to their data. And in the same way that companies are wanting to understand their customer at any time across any channel, customers are demanding to have their information available at any time across any channel. And these technology and market changes are happening faster than most organizations can change their infrastructure to respond. So data governance comes into play because it ensures that the right people are brought together to make the decisions and to determine the standards, the usage and the integration of data across projects, specific areas and lines of business. So this means it's getting the right people to the table at the right time to make the decisions around the data. At the same time, data governance has a role to play in smoothing out traditional project-based implementation processes that in fact create additional silos within the organization. So fundamentally, data governance is important because it is that practice that ensures that we manage data as the strategic value-producing asset that it is meant to be. So how do we define data governance? Well, we think about data governance as an organizing framework. It is definitely not a one-size-fits-all. And that framework guides the process to establish the strategy goals and guidelines that ensure that data is accurate, available, timely, auditable, secure, and consistent across your organization. And so because data governance links traditionally technical capabilities like data modeling with traditional business capabilities like strategy and communication, data governance really is the glue that brings together business resources and IT resources around the conversation of managing data as an asset. And because it pulls together multiple people from multiple organizations who create and use data in different ways, this is both one reason why it's complicated but also one reason why it's incredibly important to have some level of organized standard, strategy, and approach. So now we're going to go into a bit more depth regarding data governance, leveraging some of the First San Francisco frameworks that we use within our practice. So this is our data governance framework. And this framework is a comprehensive series of components needed to implement an effective data governance program. So you can think about this as a bucket list of all of the things that you should be considering when establishing a data governance program for your organization or for your enterprise. So let's just go through this briefly. So essentially, this framework, you always want to start with a strategy. So to a certain extent it's left to right, but not 100%. We recommend starting with a strategy. And that strategy is the articulation of your vision and mission, goals and objectives as it aligns to your corporate objectives. This is where you create the why and where you align it to the business objectives and goals of your organization. And you create an agreed upon set of guiding principles that will help to establish your data governance program and discipline and create meaning within the practice. Now, after you develop your strategy, then from there it's not necessarily left to right. Meaning that, as soon as you create a strategy, you also need to think about how do you communicate that strategy out to the rest of the organization and what potential change may come up as a result of the articulation of that strategy and the communication process of that strategy to the organization. So in this way, it is really and truly a series of components that you want to consider when determining how to implement data governance within your organization, not just a left to right process. Okay, so having said that, after you create the strategy and establish your goals, you need to identify who is going to be accountable for the execution of the data governance program and establishing data governance as a discipline within your organization. So that's why we have organizations as the next module. As the next module, both within this framework and as the next module within the education series. That organization structure then creates the accountability to execute on policies, processes, and standards as well as to measure and monitor the progress of the program to implement technology to support the strategy that has been identified. So many people think about data governance just as the bubble called policy, processes, and standards. So I think it's very important to understand that yes, policies, processes, and standards are governance, but if you just focus on the creation of policies, processes, and standards, it's very difficult to make your data governance discipline sticky within your organization. It's difficult to move that organization through the change necessary that is needed to have data governance be a productive practice within your organization. So that's why we've created this framework to take into consideration all of the different ways to make data governance sustainable. To ensure that you have proactively addressed the people side of the organization through a proactive change management, training, and communication plan, and that you consider how you're going to measure the success of your program, not just from a progress perspective, but measuring the impact of the data governance program on the organization. And technology we see as a great enabler to data governance. Technology isn't data governance in and of itself, but it is a great enabler to make a data governance program effective. So this is the way that we think about data governance and a framework in which to consider how data governance should be implemented within your organization. This framework will create the context for the rest of the education series around how to build out a governance program within your organization. So governance actually is associated with another practice called enterprise data management. Enterprise data management is really those activities that are executed within an organization guided by data governance to manage the data within your enterprise. So the blue bricks that you see within this enterprise data management framework, we think about as core capabilities needed to implement governance within those contexts. So to implement the guidelines and the policies and the standards that are created by the data governance program as it pertains to the different categories of data. So data governance essentially provides a wrapper around other activities and capabilities needed to manage data across the enterprise. Those capabilities are things like master data management and the capability of understanding what is a master record in your organization and how is it created, managed, and propagated across your different systems. Reference data management as well. Metadata management which includes the process of data definitions and business glossaries. So many times these are considered to be technologies, but I'd like to challenge that perspective and to have you think about the blue bricks, not as technologies, but as capabilities to manage data within your enterprise guided by the business context that's provided by data governance. And this is the link between data management and data governance which will be explored in more detail in the last module of this series. Now if we take this to the next level, really the big picture is enterprise information management. And an enterprise information management program ensures that the data as well as aggregated and derived data, also known as information, is managed consistently across the enterprise and is managed in a way to ensure that it is a corporate asset. So an EIM program or an enterprise information management program ensures that the way that information in the form of data, reporting and analytics, documentation, and other sorts of content is created and shared across an organization and externally to business partners and affiliates. So the EIM program ensures that information is consistent, effective, and efficiently managed. So that information is a value-producing asset within your organization. And of course a good EIM program leverages architecture and technology as you see as essentially the foundation of this framework to enable more effective and efficient processes within the enterprise information management program. Having said that, what is the role of technology? Well, I've seen many organizations actually start an information management program by buying different technologies to manage their information. So whether they buy a master data management hub, a data governance tool, reporting software, or whether they start to implement a big data analytics environment, that's where a lot of companies start because as we've done in Silicon Valley, we have created this technology and software industry that solves lots and lots of problems. Well, what's missing is actually other more important tools that you see identified on the right-hand side of this screen. And these are tools like business processes and capabilities and stakeholder support that ensure that the software implementation is successful and that that software is truly used and consumed in a correct way in the environment. So essentially this is a very synergistic process because of course that software also makes information management and data governance more efficient by automating some of the quality processes, by automating some of the measurement and monitoring processes, and by automating the way that you can measure the quality of your data and the way the data is used across your organization, then you can use those measurements to better communicate the value of governance to your organization, which then helps to maintain the momentum behind your data governance program. So technology and data governance are very synergistic and it's optimal to leverage both to make sure that you have the organizational buy-in and understanding of why data governance is important to the organization as well as the question for me, to the individuals, as well as the technology to support that concept. So because the goal of this education series is to establish successful data governance programs, we're going to talk about success factors a couple of times. So this is our first slide where we talk about success factors within a data governance program. And over the years, we have learned that the most important success factor in any data governance program is actually understanding why you're doing data governance in the first place. So as I mentioned on the previous slide, some of these other must-have tools like the business sponsors, the executive sponsors and support, the solid alignment between business and IT, all of these can be articulated through a clear understanding of the business drivers, business goals, and the business case for data governance. And if the organization understands why data governance is important to the organizational success, then it makes it easier to motivate people to adopt the policies, standards, and processes that are outlined in the data governance program. And that's why having a clearly understood and defined business case for data governance is critical. It establishes the fly for the company, which is the starting point and leads to establishing the what's in it for me for the rest of the organization. And that what's in it for me will help with the inevitable change that occurs when you're implementing a data governance program. Hence, another critical success factor is proactively addressing that change and creating a change management plan to address potential change issues and resistance that may come up within the organization as a result of the data governance program and the request to manage data differently than people have in the past. And of course, that change management capability will help to ensure that the people that are identified as participants in the organization understand their ownership and accountability and can be effective participants within the organization. And that organization crosses all levels from senior executives down to the people that create data and use data in their daily jobs. Now, to discuss this last bullet point, although data governance is considered to be a program and the establishment of data governance is creating a new discipline within an organization, it's ultimately implemented as a series of projects. So we have found that one of the critical success factors for sustainable governance is to be able to effectively implement projects through trained, educated, and knowledgeable resources within the organization. So effectively managing those series of projects helps make the overall program much more effective. So if data governance is so important, why is it so hard? Well, unfortunately, there's lots of internal obstacles to data governance that exacerbate some of the external market and industry pressures that are put on any company and any organization. First and foremost, data governance is hamstrung by the view of it being overhead or bureaucracy. And the resistance that comes with the perception that data governance is going to slow down the way that business is executed within the enterprise. That feeling of bureaucracy is exacerbated by a competing pressures based on time and resources when folks have a lot that they need to be doing within the organization and they feel that their priorities are constantly shifting away from data governance. Sometimes this is also exacerbated by data ownership or the lack of data ownership. So this is the concept of the data is mine and you can't use it or a flip side being the lack of ownership where nobody takes accountability for the data. So there's a negative impact by having people be too accountable and trying to prevent other people from using their data. But at the same time, if nobody is accountable for the data, then nobody is accountable for the quality of the data. A lot of times data governance is the first program that is attempted to be implemented that requires cross business unit coordination. So those organizations that don't have experience pulling together multiple business units to work in a cohesive and collaborative environment have a big challenge with implementing this. So I'm not going to go through all of these obstacles in detail, but what you can see happening here is that there are many, many personnel and people-driven issues that come up from a governance perspective. So we'll address those people issues as we continue in this module, but also as we talk about the organizational constructs in the subsequent modules. But fundamentally, this is about balancing the requirement for control around the data to make sure that it is accurate, secure, and supported with the desire to increase efficiency and reduce the effort that's associated with managing the data. And ultimately, successful data governance is balancing the requirement for control with the desire to reduce the effort. And thereby serving the goals of the business, which is to increase the productivity and reduce the effort around the data with the desire to have control around the data. Okay, so now let's turn to specific business drivers. So we did talk about some of these in the previous slides where we were talking about why is data governance important. But let's talk about some of these other categories of business drivers, lead organizations to establish a data governance program. So legal compliance and regulatory drivers have helped this industry significantly by highlighting the role of data in understanding risk and mitigating risk within an enterprise. So within those highly regulated industries like financial services and health and life sciences, we have the government to thank for pushing forward data governance as a body of knowledge and as a practice and discipline within the industry. Most of those regulations are focused on an organization's understanding of how the data is created and used within their enterprise and the ability to demonstrate and prove accountability for that data and an understanding of transparency of the data as it moves across the organization. So because of those regulatory drivers, many organizations have started on data governance programs. Now what we have found is that if that's the only reason that organizations create a data governance program, it in fact becomes a bit limiting because they will do the minimum to adhere to those regulatory drivers without looking at truly how do we manage data as a corporate asset. So those organizations that leverage the external driver of regulatory compliance with internal drivers such as executing across their corporate strategy, those organizations are most successful. So looking at how does data contribute to the ability of that organization to execute against their mission, vision, objectives, and goals is really a critical success factor for governance. So understanding that the requirement or understanding that data is a key requirement to execute on those corporate strategies is very important. Now there's other sorts of IT drivers such as just a fundamental need to increase data quality, the experience of having multiple delayed or failed implementations because there's not been the proper recognition of the data that's included in those systems implementations. Those can also drive organizations to implement data governance as well. Those sorts of industry or those sorts of information technology drivers lead many organizations to start their governance program within the IT organization themselves. And then of course, there's external industry drivers that are also pushing forward the need for data governance such as the desire for consistent and standardized information that can be shared across a supply chain is pushing consistent data governance practices in different industries. The desire to optimize customer information and customer interaction, as we mentioned before, is both an internal and external industry driver as well. So lots of reasons to, from a business perspective, to be pushing forward data governance. But ultimately, the goal is to align the business or align the benefits of data governance with the business goals and objectives of the organization. So what you will do as you go through the implementation of your data governance program is you will take a look at the typical benefits of data governance and determine why is this important to my organization and how would I measure the impact of those benefits to the organization. And this is how you would determine truly how data governance is assisting the organization to move forward and better execute their corporate objectives. So if we think just fundamentally around one of these benefits, for example, data transparency and quality, the idea is that we want to determine how does data transparency and quality improve our business. So if we better understand how data is created across multiple channels within our organization, we can better understand how those customers interact with us across multiple channels and we can react from a business perspective to optimize those customer interactions across those multiple channels. So the idea here is that you want to tie that benefit of data governance and in this case data transparency back to ultimately the business goal of, in this case, improving customer experience and optimizing your customer interactions. And in this way, by aligning to the true business drivers that we discussed in the previous slide, then you can start to understand the true impact the governance will have on your organization. Okay. So now I'm going to go through a few risks, pitfalls, and success factors. So this is where we're going to go through some lessons learned and hope to provide some guidance so that you're not necessarily learning everything the hard way as we have done over the years that we've been in business. So by learning from experience and from where other people have been challenged, you can avoid those challenges as well. So if we think about the risk of not governing data, we can think about it in the way that a company itself understands and estimates their risk. So the idea is to go through the typical risk categories that are identified and measured in any organization and understand the data component of these different risk categories and how the lack of controlled governed and high quality data would increase that risk in that category. So what I'm going to do is I'm just going to touch on each of these categories of risk and talk about how better quality data either reduces the risk in that category or helps you to better understand and measure your risk within that category. Okay. And not all categories are equally impacted by the quality of data. Some are impacted more than others and it also depends on your industry. But in general, it's important to understand the risk of not governing your data in the same terminology and same language that the company discusses risk in general. So strategic risk is understanding what is the risk of our company's ability to execute on our strategy. So how will competitors react to new product launches? How will this acquisition enable us to grow and secure market share, et cetera. And clearly there's a role for data analysis in decision making and how to address competition, how to understand merger and acquisition in any sort of corporate event that would be critical to the expansion of a market. Now another way to think about this is that there is a huge component of a merger and acquisition or bringing in another organization. It is data oriented. So therefore data has a big role in making sure that any merger or acquisition is in fact effective and actually meets the goals that you identified when you're thinking about acquiring that business in the first place. So from a strategic risk perspective there's two ways that data has an impact. One is in understanding what that risk is and also in ensuring that when there is an event that occurs from a strategic perspective that that event is managed in such a way that it does deliver the value that's expected. So financial risk is whether the company has enough cash to meet its obligations to employees, shareholders, partners, et cetera. Within the financial services industry financial risk is also understanding market risk and understanding how the markets impact the ability to execute as a financial services organization. But a great example here is that when the recession started in 2008 when a lot of the meltdown started to occur there was a big scramble for CFOs to understand their cash flow collections risk and overall financial risk as it pertains to some of the big suppliers and some of the big manufacturers that were starting to become insolvent and go bankrupt. So I spoke to multiple CFOs that were rapidly scrambling trying to determine how much is due to them from companies like General Motors and what would happen if General Motors didn't pay their bills or fulfill their commitments. And so understanding every place that General Motors is either buying from you or you're selling to General Motors is a way that data can inherently help you to understand your financial risk, your cash flow risk and your collections risk. In the financial services industry the same thing was occurring where financial institutions were scrambling to understand their exposure to Lehman Brothers. Well, Lehman Brothers isn't always represented as Lehman Brothers in all categories of transactions based on the number of affiliates, partners, joint ventures and other organizations that might be partially owned or wholly owned by Lehman Brothers. It's not necessarily represented just as Lehman Brothers as that counterparty. So it's very difficult to pull together all of the data to truly calculate the exposure to that entity as it went out of business. So this is how data can help to understand financial risk and also to react to it. Now we talked about how governance can help to execute against regulatory requirements but I'd like to also talk about another category in which improved data and well-governed data can assist to reduce regulatory risk and that is the cost it takes for an organization to adhere to different regulatory requirements and the cost it takes every single time a new regulation comes out. Well, the better infrastructure that you have around your data, the better you can respond to those new regulations as they come out and your cost to adhere to those regulations can be reduced significantly. From an operational perspective there's many ways that data contributes to increasing operational risk or reducing operational risk. So some of these things are understanding how information is created and shared and therefore is there a risk of losing information, whether that's through information that is held on an individual's laptop, information that is held in a non-secure building. But one thing that we've seen come up that isn't an obvious one to most organizations is that data actually can have an impact on employee turnover and we saw within one of our clients that they were hiring very talented and expensive MBA students into their organization but because their data was not well-governed and was not of high quality these expensive MBA students were actually spending more time doing data remediation than they were doing data analysis and as a result the turnover in certain data oriented jobs within this company was quite high and improving the data actually led to a reduction in employee turnover because those newly minted MBAs were actually able to do the analysis that they were hired to do versus just the remediation to clean up the data. So anyway, a few thoughts around the risk of not governing your data. So let's go through a few pitfalls and these come up quite a bit because folks have, as we have progressed and matured data governance as an industry we've identified some things to avoid. So the first is governing data within IT. So it's not a pitfall in and of itself because of course data needs to be governed within IT but data cannot only be governed within IT. Data needs to be governed across the entire life cycle of that data and usually data is not created in IT. It may be managed in IT but it's not created, it's not acquired by IT and it's not necessarily analyzed by IT except for maybe an equality perspective. So it's important that data is governed across the organization, not just within IT. And that kind of leads to the next one which is a pitfall of governing data in silos. This happens many times in very large organizations where the coordination is very challenging across multiple lines of business and so we will address some of that in the next module which talks about the operating model but the idea is that you want to be able to govern data consistently across your organization even if the implementation of that governance is unique to the operations of that business unit or that region. And the idea here is that you're thinking globally about your data because it's a global asset of your enterprise and you're acting locally as it pertains to your job, your business unit or your region. I'm going to jump to one of my favorite pitfalls and this is using meaningless metrics. So one of the biggest challenges around governance is metrics and measurement. And so organizations will scramble to identify something to measure so that they can be at least measuring something. So what can happen is you end up measuring things that aren't really meaningful to the business impact that you're looking for within data governance. So you're spending some time measuring your progress but you're not necessarily measuring your impact and that progress might be important to you as a data governance owner but it doesn't necessarily measure the impact to the organization. So when you're creating metrics it's important to measure impact as well as progress. The next one I'm just going to touch on is ensuring that the company culture is taken into consideration when you're implementing data governance. We found that it's very difficult to push the boulder uphill or to try and change culture via data governance. So take into consideration the norms of the organization, whether it is you consider it a positive norm or a negative norm but take that culture into consideration when you're creating your program. And then lastly, a few ideas around ensuring success. So over the years, we have found that the following factors are usually evident in a successful program. And again, I'm not going to go through all of these bullets but I'd like to highlight a few that are really critical. So the first one, creating a strategy and then following it. So this is one that I think is incredibly important, is that there is a documentation of what you are trying to accomplish with governance. And that documentation, even if it is written on the back of a napkin or whether it's a hundred-page tome of information, it really doesn't matter how big that strategy is. But it's a documented articulated statement of how data governance will benefit the organization and how you can use that strategy to rally the participation of different parts of the organization. And that's a really critical starting point because that strategy will help with all of the subsequent bullet points that you see. It will help align business and IT. It will help to define what that measurable success criteria is so that you are measuring impact, not just progress. It will help to articulate how you itemize those iterative or that iterative deployment process. It will help recruit your executive sponsors and it will help to articulate why data is important to different stakeholders across the organization. The other thing is it will give you that foundation for communication of data governance across the organization and why data governance is important to the organization. So that last bullet point of communication is absolutely critical. So to wrap this up, we're going to go back to the basics. So let's talk a little bit about design principles. And these design principles will help ensure that you don't fall into those pitfalls that we discussed and that you will also take into consideration those lessons learned or those success factors. So it's important to be clear on what you're trying to accomplish. So as we talked about the business drivers, the business goals, be clear on what you're trying to accomplish with governance and what that impact would be. And going back to thinking globally and asking locally, there should be an enterprise viewpoint around governance when you're designing your program. Even if you're implementing governance on a project or on a strategic project just to start with, understanding what the ultimate enterprise impact will be is really important. It's important also to be flexible. Businesses change. Requirements change. People change. So it's important to have a model that is changeable and flexible as the organization changes because change is inevitable. What is it? Desk tasks isn't changed. They're all inevitable. And then just to highlight, again, you see the exact same thing on the bottom, communicate, communicate, communicate. It is really important that you set up within your data governance design the process of communication. Who communicates? Who is communicated to? How frequently does that communication happen? What's the contents of that communication? So ensure that communication is incorporated into your design, not just into a series of activities and that it is staffed for, it is resourced for because it is very important to the organization. So how do we get started? So the first thing is to assess and to understand what you have. What is your current state? Where is data being governed? So even if you don't have a data governance program in those terms, every organization is governing data in some way. So it's important to understand what that is because you don't want to reinvent the wheel and you don't want to redo something that is already working. So that was one of the design principles that I didn't highlight, but the idea is that you want to leverage work that has already been done in the past. So assess your current state, inventory current data governance practices, inventory your data management technology. What do you have? Who is involved with it? Who are the stakeholders? How do you align the goals of data governance and create that business case? So all of that initial assessment in the current state is a very foundational activity to ensure that you're building upon that foundation. And this is a good idea to start creating some baseline metrics at this point as you're assessing your current state. Now the next step is, okay, if we have assessed our current state, we've created a business case, we've started to articulate what that future state will look like. And if we can articulate what that future state will look like, then we can create a map and a plan and a strategy to execute against that future state. So this is where it's important to create that roadmap that will guide you to that future state and that it will guide you to execute that strategy. That roadmap will help to articulate how you are addressing all of the different components of the data governance framework that we introduced in the very beginning of this module and presentation. And that roadmap will be your guide to implement data governance across your organization. So roadmaps can change, but at least you have a point of articulation of what is going to happen, who is going to do it, so that you can start to execute and implement governance according to your plan and not in an ad hoc, unorganized, uncoordinated way. So this is the end of module one and we appreciate you going through this module with us. We encourage you to take the assessment questions and then join us again for module two, where we will drill down into creating a data governance operating model, which is the core of establishing a data governance organization. Thank you very much and enjoy the rest of your day.