Data continues to become more and more important. With this growth, there has been a corresponding need for standardizing and managing. In this post, we will look at master data and how one can go about managing it.
Definition
Master data is a uniform set of data that is used throughout an organization. By uniform, it means that this data is exactly the same wherever it appears in any data set. This is highly important because it is natural for data to change a little as people use it or if it is merged and edited in various stages of the workflow. Master data is so important and fundamental that it must remain unchanged for the sake of consistency when different departments within an organization need to integrate data.
Therefore, master data management is the process of protecting master data from the changes that can happen from people and systems interacting with data. Unfortunately, preserving data is not an easy task and at times this can be complex and difficult.
Master Data Management Forms
There are different forms or ways to develop master data. The analytical approach feeds whatever master data an organization has into a data warehouse where it can be referred to as needed. The operational approach involves master data in the core business or organizational systems. Essentially, the difference between these two approaches is at what level of granularity they are implemented. Analytical is across an organization while operational is within a sub-unit of the organization.
Whichever method is used there are several ways that the approach is implemented. A registry process involves creating a unified master data source with making any changes to local systems. This means there are two different systems which mean that people need to be aware of when to refer to the registry.
Consolidation is another way and involves updating the registry master data whenever the local system is updated. Lastly, the transaction method is the opposite and involves the local system being updated whenever the registry is.
Steps
The steps to selecting and standardizing master data are explained below Step one involves selecting what is considered master data. This will vary from organization to organization and will involve some disagreeing and negotiation. The same applies to step two which is agreed on data standards and the master data approach. Examples of things that involve data standards can include capitalization of text, number of decimals, number of digits, maximum text length, abbreviations, etc. All these must be worked out together. Should states be abbreviated or spelled out fully? Should phone numbers have dashes in them? These are just some of the challenges to address.
Step three involves deploying the software to find and standardize the master data. This can be done manually and this happens in smaller organizations but for larger organizations, this is the only practical way to do this. Step 5 is the cleansing of the data which can include dealing with duplicates. Once all of this is completed it is now appropriate to use the master data.
The Team
Most projects require a team effort and master data management is no exception. Often you will want a manager who oversees the project. Another person who may be involved is a master data specialist who maintains the system. Data stewards are generally involved as they are the ones most familiar with whatever data they are responsible for. In addition, you may need leadership sponsors and stakeholders involved as well particularly when picking master data and assigning data standards.
Conclusion
Master data is a critical component of many organizations which means it must be managed and controlled as well. Some practical ways to address this have been shared here. However, the best way to approach this will vary from one organization to another.