A major US based global retailer has used Collibra Data Governance Center to govern their use of item attributes across the organization. The digitization of their business has created a tremendous explosion in the number, type, and variation of attributes on the items they sell. This attribute explosion, a complex system and data landscape, and new requirements from big data systems caused the time to add new items to balloon to 6 months. This created a significant constraint on their ability to introduce novel items in their stores, and respond to changing demand and tastes.
Previously, there had been efforts to harmonize this data by using master data management techniques and programs, but that effort was challenged by varying syntax, semantics and policies used in different parts of the business. The separation between the store systems and e-commerce systems was particularly challenging, because they were run by separate teams in different facilities thousands of miles apart.
All of the information and processes were tracked using spreadsheets and word documents, and were hosted in a SharePoint site. Because the information was not easily accessed, and was not tied to specific tasks, accomplishments and day-to-day activities, it was not useful. Also, it quickly became outdated and therefore even when it was found was not always trustworthy. If the governance process required changes to the information management and data quality systems, these were handled by manually created trouble tickets in the IT ticketing system.
Collibra Data Governance Center was used as the foundation of a data governance solution. This solution linked their data stewards, data owners, and subject matter experts into a collaborative process-based environment. They were able to harmonize the data and create rules and policies that maintain its quality.
The result was that there was a clear understanding of what the data was, what it meant, and how it should be used in the different parts of the organization and in different systems. Most critically, the process of adding new attributes, and new quality rules for those attribute values has been systematized and automated, so that it happens when needed. The processes and responsibilities are clear, and the activities are orchestrated so that they are completed rapidly, efficiently and completely. The effect of governing item attributes was so successful that this retailer began expanding the governance effort. The Governance team can create new item attributes with direct integration to the big data systems. This eliminates some of the IT manual work. In addition, the Collibra Data Governance Center is integrated with the IT ticketing system, so changes that require IT execution are automatically ticketed and the IT team is provided with all of the relevant information about the change. In addition, both teams (IT and Data) have visibility into the status of any outstanding requests.
The first expansion was to include vendor attributes as well as item attributes. Vendor attributes have a very similar set of issues and complexities to item attributes, and similar processes (but different people) are involved in defining and creating rules for those attributes. While the time delay involved in setting up vendors is not quite as severe, this organization is reducing a process that normally takes months down to weeks, and possibly even days in many cases.
The second expansion is to use the (now precisely defined) attributes and rules to improve the quality of the data provided by suppliers about these products. The current data quality is measured through a variety of data scanning tools combined with operational metrics. Collibra is fully integrated into these tools, and the data quality measures are rolled up into scorecards using Collibra Data Governance Center. The scorecards offer complete visibility into the data quality picture, including measures from all sources, rollups of composite data quality, historical trends, and stoplighting that enables users to easily identify problem areas. In addition, when data quality is too far out of tolerance, data issues are automatically surfaced with Collibra and routed to the appropriate individuals for resolution. Suppliers are given their standing in terms of their data quality relative to their category peers, and are informed of any issues that could cause errors immediately.
Finally, for both the attribute definition and data quality issue remediation this retailer is involving the suppliers directly in the process. Mangers of the data at the suppliers are active users of the Collibra system, either directly through the web interface or via the interactive email integration. This dramatically streamlines the process of collaborating and resolving any quality or definitional issues, and insures that the overall quality of the data is continuously improving, rather than degrading. Ultimately, this creates a level of trust in the data, so that users are more confident about relying on it to drive decision making within the organization.
The solution was up and running within 10 days, and was fully operational within a month. It supports over 500 users from across various teams such as supplier management, product teams , and business analytics reporting.
Overall, this retailer found that taking an aggressive approach to governing their item and vendor data has shortened the time it takes to make system changes, improved data quality, created trust in the data, and engaged a collaborative process with its suppliers for issue resolution. All of these processes were powered using the Collibra Data Governance Center.