I mentioned in the earlier columns of this series (part 1, part 2) that organizations are focusing their data governance programs on improving their data for analytical purposes. I may have even mentioned that organizations have left their data unattended for years, leading to less than desired confidence in their data. The solution to this “bad data” problem begins with improving the quality of data that is defined, produced, and used within their organization. This column focuses on data usage.
As I mentioned in the previous columns, many organizations focus on achieving “good data” through improvements in how they define and produce data. Still, every organization wants to be able to predict customer behavior, improve efficiency, and effectiveness of their supply chain, reduce production costs, … or whatever the business USE of “good data” may be. It’s All in the Data.
This column addresses what it takes to achieve “good data.” For this column, I will focus on the third area where organizations can simply and logically break down the activities that are required to achieve “good data.”
The three activities include:
Improving Data Usage
Begin with Data Users’ Understanding
Leadership hates it when they ask a simple question and they get … a simple answer. In fact, they may get multiple simple answers to the same question. Honestly, that drives leadership crazy. When that happens, they have many questions that need to be answered by the people that produced the different results, including:
- Why are the results different?
- What data did you use to get your results?
- Which result is the “right” answer?
- Why is our data so different from one place to the next?
- What are we going to do to prevent this problem from occurring again?
These are all great questions. The answers to these questions can be used to get leadership to pay attention to issues that your organization has with their data. However, this column does not address using a situation like this to highlight the deficiencies for leadership to pay closer attention to governing the data as a valued asset.
The reason that the results are different from one another typically comes down to the data users’ understanding of the data. That understanding can include the definition of the data, where to go to find the “golden record” source for this type of question, how pieces of data have been manipulated, where the data came from, and the confidence that the organization should have in the data they use to answer these types of questions. Without complete (and identical) understanding of the data, it is very possible that two people will receive different answers when using the identical data sources.
The bottom line is that the understanding of the data depends on the quality, value, and availability of data documentation (metadata) to align data users’ understanding across the organization.
Apply Accountability for Data Usage
Another facet of data usage is the need to hold people accountable for how they use the data. This is easiest to explain in terms of protecting sensitive data – like personally identifiable information (PII), personal health information (PHI), or intellectual property (IP).
In order to apply accountability for data usage, it is obvious that the people using the data must understand how the data is classified (for example – checking if the data is classified, sensitive, public) and how data categorized each way is allowed to be handled and shared.
Holding people accountable for how they use data requires that the rules associated with handling data must be thoroughly documented. These handling rules include rules for how people can print data, send data via email, send data via interoffice mail or through electronic interfaces, secure data in their offices, desks or cabinets … as a short sample of ways data must be handled.
The truth is that accountability for data usage can also require regulatory, compliance, and business rule understanding of the data. The bottom line here is that people must be educated and continuously reminded of how classified data can be handled in order for an organization to apply data governance this way and apply formal accountability for how data is used.
Assure Data Usage Quality Control
It requires a specific and concentrated effort to completely document the handling rules, the business rules, and the compliance and regulatory rules. It takes resources and an effort to make these rules available to people that use your data. It takes even more resources to educate and train all of the people in the organization on how to access and follow the rules. It also requires resources to monitor how data is used, report, and follow up when the rules have been broken and take corrective action. But what choice do you really have … if you are to provide quality control for how data is used?
Quality control is often thought of as a front end or production process. When it comes to quality control for the usage of data, that control (or in other terms – governance) must also be applied to the definition and production of data that I wrote about in the prior columns.
As I mentioned earlier, this column wraps up a three-part column that focused on assisting organizations to achieve “good data.” Using data to improve decision-making and analytical capabilities is a result of governing all three of the definition, production and usage of the data. Remember … It’s … All in the Data.