Data quality is an important piece to data governance and thus important for a higher education institution. During our various webinars and blog posts we have mentioned several thoughts that we are sharing in this blog post to improve data quality.
Not in priority order, here are our thoughts:
- You need a data governance and data quality culture in place at your institution. This means a focus on the people, processes and content that is necessary. Perceived data quality erodes trust in data regardless of whether it is real or not. Building trust in data is more important (in some ways) than solving the issues. Evaluate whether you have this culture in place. If not, work to create this culture.
- If you don’t have one, then set up a data quality program at your institution. Read our blog post on data quality programs.
- Start simple. Again, if you haven’t done so, create a simple knowledge base of quality rules, document “first step” processes such as a quality issue resolution process and start with one data steward or one business area. People need to know who to contact when they see a data quality issue.
- People are a key component of data governance and data quality. Make sure you on-board new employees on the importance of data governance and data quality as well as provide ongoing training of all employees including data stewards. Read our blog post regarding on-boarding.
- Processes are a key component of data governance and data quality. Have in place processes for data issue resolution, quality rule management, root cause analysis, data quality communication and assessing of data quality.
- As in most programs, communication is key for success. Make sure the appropriate people are communicated with about changes and issues related to data quality. Communication builds trust in your data.
- If you haven’t done so, do a data quality assessment. It is important to communicate the assessment results to others. To learn more about assessments and the steps when assessing data check out this blog post on the subject.
- Make sure you have one source for tracking your data quality issues and records the resolutions for these issues. To find out more about steps in the data quality issue process check out this blog post.
- Root cause analysis and resolution is more important than cleanup of specific data. Don’t just put you finger in the dike, fix the main problem (integration issue, missing rule, etc.).
- Create a Just-in-Time (JIT) data quality philosophy where you quickly fix errors, quickly move issues into rules, quickly document the rules, quickly do root cause analysis and resolve them so don’t happen again. These JIT actions enhances trust in your data.
- Have in place the content in place for data quality including quality rules, quality assessments and quality issues. Make sure that this content is easily accessible. Content is a key component of data quality. Read our blog post regarding data quality content.
- Best time to add content for data quality is when data quality rules are identified such as during report and integration development. Turn data quality issues into rules for documentation and on-going monitoring.
- Managing reference data as an enterprise is important. Your drop-down values must be consistent throughout the institution or you will have data quality issues.
- Document your data quality rules and store the rules in a catalog that is easily accessed. Once a quality rule is in place, processes or software changes can be created to ensure that data follows the rule. Learn more by reading the data quality rule blog post.
- Data quality is not just database integrity which are technical in nature. Business rule issues are just as important, if not more, as it builds trust.
IData has a solution, the Data Cookbook, that can aid the employees and the institution in its data governance and data quality initiatives. IData also has experts that can assist with data governance, reporting, integration and other technology services on an as needed basis. Feel free to contact us and let us know how we can assist.
(image credit StockSnap_RHQN7LFL5D_dataqualitythoughts_BP #1078)