Data Governance Best Practices in Focus: Gaining Clarity and Learning from Data Governance Experts

As someone who’s worn glasses his whole life, I’m all too aware that the phrase “20/20” means that you have perfect vision. For many, the year 2020 also provides a new opportunity for clarity in your professional career. This time of year, there are countless posts online about how to best take advantage of the new decade, ways you can work on yourself, and tips for setting goals for the year ahead. If you’re like me, then learning from other’s experiences—whether in books, in person at events, or online—becomes a cornerstone of growth. As you embark on this new year, I wanted to let you know about a new online series of webinars—the Data Empowerment Expert series—which will feature the best of the best practices out there for data governance professionals, all with the goal of helping you to have perfect vision in 2020 and beyond.

Each month, I’ll sit with different Data Empowerment Experts, and discuss how today’s data governance leaders are empowering their organizations with trusted data to deliver business value. This month’s webinar features L.A. Care Health Plan, the largest public health plan in the United States.

As a preview for each episode in the series, I’ll also post a Q&A-style blog with each webinar’s guests. And although I’ll ask each expert the same questions for each blog, their viewpoints will be uniquely their own and will be representative of this diverse group of governance leaders in 2020.

The first two speakers in the Data Empowerment Expert series will be Paul Keller and Gary Georges from L.A. Care Health Plan. Below is our question and answers on the state of data governance in 2020.

Question: How do you choose where to start?

Answer: L.A. Care performed an enterprise-wide risk analysis, which considered 72 separate risks. For L.A. Care, high-quality provider data and network associations are fundamental to our business model; low quality provider data and network associations inhibits L.A. Care from optimally managing our provider network. This, not surprisingly, aligns with national statistics where provider data is less than 30% accurate. So, provider data was identified as our number one risk for data governance to tackle.

Question: How do you maintain momentum?

Answer: A committed executive sponsor is key because instantiating a data quality and data governance program causes significant change ripples through the organization. Because data governance is a horizontal function, it can only have authority through executive sponsorship. Generally, organizations resist change—the longer the status quo has existed, the greater the resistance can be. The executive sponsor ensures that the organization pushes to value through what will be inevitably difficult changes to make. Everything comes after that.

Question: How do you address scale?

Answer: In order to address scale, it’s important to start in a single domain before trying to scale to the entire enterprise, as prescribed by best practices. At L.A. Care, data governance tied itself to the MDM program within the provider data domain. The idea is to build data governance policies, processes, procedures, standards, service levels, and governance org structures (i.e., DG Council, DQ Council, Stewardship Network) for the provider domain and then take those learning to the next domain (for L.A. Care, that was the member domain). Start small, create and test all the data governance collateral, and learn from the inevitable mistakes made. The second chosen domain can then be implemented much faster, easier, and with higher quality. Don’t try to boil the ocean.

Question: How do you measure success?

Answer: For provider data, we created 60 reference data standards as well as 11 primary source validations. In addition, developed over 2,200 data quality rules to measure data quality across 10 different data quality dimensions. Data that did not meet quality standards was sent back to the source (e.g., managed care organizations) for remediation. Based on standards, initial data quality from 44 delegated health plans (82K records each month) was very low (10-15% overall.) Through data quality standards, cleansing, and remediating the data, we were able to improve data quality a stated below:

  • Intrinsic dimensions: Accuracy (>95%), Semantic (100%), Structure (>95%), Lineage (TBD)
  • Contextual dimensions: Completeness (>95%), Reasonableness (~60%), Currency (85%), Timeliness (TBD), Identifiably (98%), Consistency (TBD)

Question: How has data privacy been a factor? 

Answer: As a second initiative, data governance took on data security, which had multiple C-suite advocates and is being led by the CIO. Within this program, we are performing a potential impact assessment of a breach in confidentiality, integrity, and availability. In addition, we are conducting a data sensitivity assessment. The results of both assessments guide how both electronically store data and non-electronically stored data should be handled, by whom, and in what circumstances, using either Role-Based Access Control (RBAC) or, better yet, Attribute-Based Access Control (ABAC). Given the highly regulated environment within which health plans operate, data privacy is undoubtedly one of the top risks faced by these organizations.

Question: How do you empower the business to drive value?

Answer: Our strategy is to utilize Informatica’s Enterprise Data Catalog (EDC) and Axon Data Governance tool set in order to identify and catalog business and technical metadata within the provider data domain and information security. We can publish the approved reference data standards in the EDC, integrated with Axon’s business context (e.g., definition, lineage, regulatory references, and L.A. Care policy references), while identifying and eliminating current rogue use of non-approved data.

To hear more from Paul and Gary as they share L.A. Care health plan’s journey with data governance, please register to join us on January 28, 2020.

I’m thrilled to work with each of these leaders to help tell their story. To learn about other speakers and register for future segments that will be aired in February and beyond, please check out our Data Empowerment Expert series page.

Comments