Advertisement

Data Governance Best Practices: Five Industry Experts Discuss Their Viewpoints

By on

data governanceAt the Enterprise Data World 2017 Conference, a panel of Data Governance and Data Management industry experts shared their wisdom, offering tips, ideas, and best practices. The panel discussion was titled “You CAN Be Successful in Data Governance Best Practice Tips and Takeaways.”

The transcript below is a selection of some of the highlights of the panel discussion. To watch the entire dialogue, see the full video at the bottom. Each of the panelists answered specific questions presented by John Ladley, the Panel Moderator, on topics including: Starting a Data Governance program, how to sell Data Governance to your organization, what problems could arise when setting up a new program, the role of Data Stewards, and others.

The panel included:

First Day in a Data Governance Job: How to Get Started

Ladley (Moderator): “Let’s pretend that the audience is made up of new people who are now working for you. Take a minute and tell them the most important thing, in your mind, that will get them off of ground zero.”

Nicosia: “Spend time planning what you’re going to do. Don’t run off and solve problems right out of the gate, because once you do that, you’re going to go down the Alice-rabbit-hole and that’s all you will do is solve problems.”

Algmin: “Strategy is important, and doing the planning and engagement is important,” Algmin said, but he cautioned about talking about data “instead of having meaningful business outcomes.”

Delgado: “Focusing on the business process and the business challenge.” And rather than spending most of your time with the executive team working on buy-in, “Get to know the folks that are actually managing the business and processes, that are inputting and outputting that data. Ultimately that data is a result of those business processes and we often forget the folks that are actually in the trenches.”

Zagoudis: “The first thing that I do is try to figure out who really has the vested interest.” He suggests reading the annual report to understand the risks that are “significant enough to be publicly disclosed.” Then look for any audit findings, SEC findings or regulator findings that are driving the motivation to move forward. “If you then find out which executives own those issues, that gets you some sponsors,” he said. Finding people who have the interest, the knowledge, and “passion for the data,” is the second step. “They’re not going to see Data Governance as part of the problem, they’re going to see it as part of the solution.” Follow those steps and, “the program will take off on its own.”

What Does a Data Steward Do in Your Organization?

Nicosia: The basic purpose and responsibilities are to ensure the integrity, the quality, and the accessibility of the data, including defining the language used, he said. Nicosia’s job is to give them the tools and the processes to help them do that as efficiently and effectively as possible.

Algmin: Defining certain qualities for the person filing the Data Steward position is essential, because being able to understand data is not the only required skill, said Algmin. “You need to look for people who have good communication skills. They can problem solve. They’re comfortable with ambiguity,” he remarked. “Having some thought about who would be the right person to fill that role is very important.”

Delgado: At Citrix, a group of ten part-time Data Stewards meet every two weeks to discuss Data Governance from an operational standpoint, said Delgado. “There will be Tactical Stewards embedded in each of the functions that just kind of live and breathe and know the data but we also have the concept of a more ‘leader-level’ Data Steward that we’re dedicating from a customer perspective,” he said.

Zagoudis: Data Stewards must be subject matter experts (SMEs), which is why it’s hard to hire someone from outside to do the job, said Zagoudis. “I expect the Data Steward to facilitate the quality corrections to data, facilitate the awareness of the source and use of data, know who’s using their data, know where it comes from, and ultimately make certain that people that are consuming that data are communicated with on any changes.” He considers the Data Steward “the stakeholder of that data.”

Ladley (Moderator): “I’m going to add something to that because I’ve run into this a lot. There’s a rush to appoint stewards. A rush to say, ‘I have to have this role,’ without defining the role. So, forget the title. Forget that you have to appoint someone called a steward. It could be a custodian. It could be a data guru, whatever. But there has to be accountability, responsibility, and control.”

Share Your Elevator Speech

Ladley asked each panelist for an ‘Elevator Speech” they would deliver to their CEO about the need for their program.

Nicosia: “We just came out of the operating plan review and I noticed that the assets and the flows for the last three quarters have been trending down. Do you realize that you are missing out on millions and millions of revenue opportunity because we don’t know how many customers we have? Do you know how many customers we have? We have 5 million. Is that the right number? If you go ask somebody else, we have 4 million. If you  ask somebody else, we have one million. So not knowing how many customers we have is losing us revenue, because we can’t sell products to those people. So, if you’re in for losing revenue, then let’s not focus on the data. If you want to generate more revenue, then let’s focus on the data.”

Delgado: “Recently we had a meeting where the guy who sells your product and the lady who makes it provided you different metrics. I think that’s a bit of a challenge for us in understanding how we’re going to measure the business . . . [and] I think it’s very important that all the executive teams that you’re supporting talk the same language and bring the right numbers to you so that you can make better decisions. I think that Data Governance can help with that. There are some processes that help us bring different functions together to make sure that they have common definitions, and that ultimately, they’ll provide you the right metrics to drive your business. What do you think about that?”

Zagoudis: “Mr/Ms. CEO, I notice that you’ve got 950 spreadsheets that are currently used for financial reporting. You realize that there is a high probability of error in those spreadsheets. You realize that we’ve got some findings out there that we’ve got to address. We need you to work with us to get sponsorship to bring a group of people together to tackle these issues, solve your chronic spreadsheet problems, and move forward on a long-term architectural vision . . . [and] Work with us and we’ll help solve that problem.”

Algmin: “Mr. CEO, did you know in the last quarter, 12 percent of our revenue growth came from people at hotels in Atlanta, and that those people were always at Data Governance conferences? What we found was at those Data Governance conferences the average aggregate spend by those folks while in Atlanta was 90 percent more than a typical conference in Atlanta, and when they went back to their office, after a six-month delay, we found their representative companies had increased spend with our firm by 400 percent. Maybe next year we should get a booth.”

What Doesn’t Work for a Data Governance Program?

 Algmin: When organizations are starting out with Data Governance, Algmin said they often decide to have once-a-month meetings to talk about data with people they think will be able to help them with implementation. “It’s a well-intentioned idea, but the problem is that there’s no connection to benefits for them.” He said that over time the meetings become less productive. “Instead of thinking about who could help our Data Governance effort, think about who can benefit most from this,” and most quickly, and make sure they at the center of the equation instead.

Delgado: In response to the idea of piggy-backing a Data Governance program on an already funded corporate project, Delgado said: If you need the funding, definitely leverage that project, but don’t necessarily make it a deliverable of a project that’s going to end.” His experience with project-focused funding was that everybody expected it to have an end date. “Don’t make it a deliverable of a funded project that ends and might lose momentum.” If you have to use project based funding, set the expectation: “This is something that’s going to persist beyond the project. Do we have your buy-in?’

Zagoudis: Theoretical best practices imposed by consultants don’t work, said Zagoudis, but “evidence of a solved problem” does. When consultants come in with org charts and job descriptions talking about efficiency, “Fear starts to come through the organization,” he said. People start to wonder, “Well, what about my job?” or “You’re taking away my spreadsheets. I’ve spent my whole life with spreadsheets.” He said it’s essential to use Data Governance to target a specific problem that needs to be solved, and to communicate with the people who care most about that problem.  He remarked that he’s seen a lot of work go into programs that did not continue after the consultant left. “So now when I go in, I won’t even do some work unless I’ve got a specific problem that we’re solving. Get the people that care about the problem, and all of a sudden there’s tangible results.”

Nicosia: “I want to put a caveat on Michael [Delgado]’s last comments. Projects are a good way to deliver Data Governance if you actually have Data Stewards and people that are actually focused on it first, because those are the people who should be involved in the projects.” Nicosia has used the piggy-backing process with some success.  “We embed practices around Data Governance and Data Management within our projects because we’re not funded.”

Their “Aha” Moment

Ladley: “When did you know you had buy-in?”

Nicosia: In the process of implementing what they call ‘Governance Practices,’ Nicosia’s team was working with an area inside their finance department that had difficulty seeing the value of Data Governance.

“The leadership of that area just could not wrap their head around it. They’re like, ‘No we already do that stuff today. We do hundreds of hours of spreadsheet manipulation and we do the same Data Quality and all that stuff,’” he said. After three months, they were able to convince the leadership that it was a good thing. “We can always talk about the change curve – we spent a lot of time in the valley of that curve. We just showed them again what they’d get at the outside and one day it just clicked. ‘Oh, so we’re going to get this and I don’t have to spend a hundred hours a month doing this?’ and we’re like, ‘no, you can spend five minutes.’ ‘I get it.’ And all of a sudden it was rapid movement out of the change curve.”

Algmin: “My favorite story is when I was a consultant with the Chicago Transit Authority, and they were having problems with their fare system. If they wanted to get a daily report of who rode the bus and who rode the train – for one month – just two numbers, 30 days – it would take 42 minutes, if it finished at all,” he said.

“Being a data person, I said, well that’s unacceptable, we need to change that.” Algmin came up with a solution and was able to present it to the CEO, who said, “That’s amazing. You can do whatever you want.” Getting that buy-in “paved the way for us to do so many things with . . . [and] a technology organization that had no interest in innovating anything,” he said. “It led to Data Governance, it lead to new innovations, it led to new technologies, all those things, but it was because I was able to push through and leapfrog to anchor in at that executive level.”

Delgado: “The one ‘Aha!’ moment for me was with standardized reporting.” To indicate that a report had been created using standardized Data Governance processes, Delgado’s team put a small watermark on all approved reports. “One of the executives sent me a note that said, ‘My team member shared this report with me but I didn’t see the watermark, can I use this?’” he said. That was the turning point where “all of a sudden the executives are demanding governance, which in that case was ‘Where’s my little stamp?’”

Zagoudis: “If you are an owner, steward, delegate, or consumer of data, you’re a registered stakeholder of that class of data. It’s terminology that I use constantly in my work.” Zagoudis recently received a text from a client that asked, “Can you run a query of registered consumers of product code? I want to get everybody together to talk about changes we’re proposing.” That ‘simple little text’ “made me realize that they got it. They pulled the group together, they talked about it, and they were grateful that they knew what they were going to do ahead of time.”

Check out Enterprise Data World at www.enterprisedataworld.com


enterprise data world

 

Here is the video of the Enterprise Data World 2017 Presentation:

 

 

 

Photo Credit: Nata-Lia/Shutterstock.com

Leave a Reply