Data Governance Success: How to Build a 360º Perspective

By   /  October 13, 2016  /  No Comments

ad_dg360_092216Jodi Morton is Vice President of Single Family Data Governance and Management at Freddie Mac, but “Data Governance Change Agent” might be a more accurate title. Citing from an outside auditor familiar with the company, Morton’s new Data Governance program “got more done in 18 months than in the previous ten years.”  Morton spoke at the DATAVERSITY® Enterprise Data World 2016 Conference, outlining the steps she took to build a team, craft a plan, and make a culture shift that created ownership – and even excitement –  at all levels.

Empower Your CDO

To start, Morton says it’s essential to hire a Chief Data Officer who is passionate about data and knows your company’s culture. For the best chance at success, she says the CDO needs to be at the leadership table, and in the business side, not in IT:

“We had tried to do [Data Governance] at Freddie Mac and we’d get some momentum, and things would fizzle. The business had to decide it was their problem, and decide they wanted to fix it in order for it to be successful.”

Make the CDO a direct report to leadership, so the department stands up on its own – not buried down in the org chart. “It’s important to give credibility to the investment we’re making; not only the department, but the work we’re doing,” she says.

The Right Scope

Morton says it’s also important to get clarity about ownership of assets. Each product and process at Freddie Mac has a business owner who drives cultural change and ensures decisions are made with input from stakeholders. As CDO, Morton now owns of Data Governance, Metadata, and all shared information assets of the business. She says that owning assets that were previously scattered about the company allows her to coordinate centrally, so instead of teams trying to solve for themselves in silos, she solves for them across the business. Morton says she has also been able to use economies of scale to negotiate deals with vendors, finding extra capacity as a result.

Making an Effective Plan: The First 90 Days

Assembling a Team

Morton started by creating a three-person team from inside the company, each member with a needed skillset. She brought in a finance specialist with experience with data, government risk, and controls. She hired an Analytics/BI specialist with experience building data marts, and a third person who was a modeler.

A Complete Perspective

To determine their targets, Morton asked vendors to connect them with others doing similar projects, as well as researching what other financial companies were doing around Data Governance.  She also studied reports from consulting firms who had previously worked on solving the “data problem” at Freddie Mac.

Make an Honest Assessment of Organizational Readiness

“Freddie Mac has been in crisis since about 2002,” she says. Morton was the CFO during the credit crisis and since that time, the organization has come from “just surviving with no time to think strategically,” to a time where they are “no longer distracted by all the ‘have to do’s,’ but are able to focus on the ‘want to do’s,’” she says, and “you can see the change in people.” She says they still have some things to do, but the crisis is mostly behind them now.

To “address the real problems, not just our perception of the problems,” Morton did an immediate re-launch of Freddie Mac’s Data Governance stakeholder group. She started with key players meeting monthly, using an “interview and listen” process with 40 people, asking about their history, their challenges with data, and what they needed to be successful. She said that interest in stakeholder meetings was so high that some had to be turned away. They continue to meet quarterly, the makeup of the stakeholder group changing each time so everyone gets to attend at least one interview a year. She planned for dissent as well:

“We just didn’t talk to the people who we knew would be advocates for the work we were going to do. We also picked the people we knew that would be the ‘haters.’ And those are the hard conversations, but if we can win over some of the haters, that really changes how things go. Some of them we’re still working on, but we made a very conscious decision that we could not shy away from the haters.”

Components of the Plan

She then created a framework, which became the Data Governance business plan for the next year:

  • Focus on realistic, high-level achievable goals: She says, “You can’t afford to fail on these”
  • Make your process transparent to stakeholder: Transparency builds trust
  • Focus on building a foundation, freezing non-essential initiatives: This helps to minimize distractions
  • Identify what you need to have good data capability: For the business this includes everything you do should fit inside that framework
  • Do a self-assessment:  Ask “How far are we from these expectations? What do we need to meet them?” She included personnel needs in the assessment, writing job descriptions, and defining targets for the next 12 months.  She emphasized making FTE for Data Governance a priority: “You need to have a team whose full-time job is data”

Endorsement from Stakeholders

After creating a draft framework, she went to the Data Governance committee and asked for sign off on the plan. “This is the body of work we think we need to undertake” this year. She asked, “Will you endorse this? Is this the right body of work? Will it make a difference if we do all these things?” All team members voted, and minutes with each member’s vote were published, so there was “no opportunity for convenient amnesia,” she says.

Communication Started

When they set a clear and realistic near-term vision for every major area, she said it was important to share what they’d accomplished. She published milestones they’d reached monthly, increasing visibility with co-workers and stakeholders.

Supporting Compliance

Once a year, the leadership team and managers at Freddie Mac do a risk assessment, assessing all processes as high, moderate, or low risk. For the first year, Morton used risk level as a metric for prioritizing each department’s work on the plan:

“We knew whether [each high-risk process was] in compliance with the standard. Where they weren’t in compliance, that process owner had to come up with a compliance plan. We walked them through it, rather than just handing them the standards [so] each of those business owners has a memo that basically takes them through [the standards] and says ‘I am’ or ‘I am not in compliance’ with these things.”

Four Concepts That Drive Business Value

Throughout the process, she asked, “Where are the gaps?” so all concerned could see the value of her team’s activity to the overall success of the business. In interviews with stakeholders, four recurring concepts emerged:

  • Empower: Provide tools so users can be more independent
  • Simplify: Start with one source that reconciles all truth, build it once
  • Reuse: Reuse it for many purposes. Retire old data
  • Control: Create a new framework to guide data producers on the proper standard of care for the level of data they use, on the premise that “not all data is created equal”

Morton says, “If the business case cannot tie back to one of these four concepts then we have to ask ourselves why are we doing it? The reality is most things tie back to at least one, many times all four.”

Communication is Critical for Cultural Transformation

She now has one full-time position dedicated to communication and stressed the importance of making the process engaging. She has a blog, posting every three weeks, and publishes regular articles in the company newsletter. At six months, the team had a Spring Fair to focus on branding the four concepts. The fair gave employees an opportunity to share what they were doing, have booths, and learn more about Data Governance. Morton’s team gave out “I heart Data’ buttons and she said that people across the company are still wearing them – a surprising development, considering previous Data Governance initiatives were not met with much enthusiasm.

Once Established, Move to Active Governance

After the plan is in place, Morton says she continues to inform leadership about critical topics, provide clarity about what decisions are needed from them, and encourage participation. “We give them intentional power and decision making rather than just listening. We get better feedback.” By tracking and publishing votes, she says, all players have to read and give feedback. She also actively engages mid-level staff to champion the cause, increase adoption, and cascade communication.

Refresh

Morton anchored her strategy in the priorities of the business by discussing each priority area in relationship to its data needs. As an example, she said that one of Freddie Mac’s focus areas is customer excellence, so she asked what data-related business needs there were in customer excellence, and built fulfillment of those needs into the plan.

Morton says that every year these business priorities are refreshed and updated, and her team asks again “What are the data-related business needs for this priority area?” changing her work document to ensure those needs get met.

Morton’s Data Governance fair was such a hot commodity that it’s now an annual event. Calling her co-workers, “the most engaged and passionate group I’ve ever seen,” they may soon be lining up for Data Lineage Twister, each proudly wearing an I Heart Data button.

 

Register for the Enterprise Data World 2017 Conference Today (in Atlanta, Georgia)

EDW

Here is the video of the Enterprise Data World 2016 Presentation:

 

About the author

Amber Lee Dennis is a freelance writer, web geek and proprietor of Chicken Little Ink, a company that helps teeny tiny companies make friends with their marketing. She has a BA in English, an MA in Arts Administration and has been getting geeky with computers in some capacity since 1985.

You might also like...

Predictive Analytics: Giving Smart Manufacturers an Edge

Read More →