Achieving Data Quality by Building Trust in Data
Our hospitality client has been running a business intelligence environment for over 8 years with about 1,000 end users and supports operational reporting across multiple functional areas.
Data quality issues plagued the system and caused high levels of distrust with the information being presented amongst the users and management.
The dilemma that we faced with our client is that the issues with the data were mainly attributed to problems with source systems and/or procedures currently in place. The client wanted the issues fixed but was not willing to take on the change/disruption to the business to fix the issues at the source. The challenge was to come up with an incremental solution that could build credibility and not disrupt the current state of the business.
Smartbridge built an interim solution with the client that identifies data quality issues right at the time of the load. We implemented exception reports and proactive alert reports that gave visibility to the impacted metrics to the users.
This solution didn’t necessarily resolve the data issues but it gave the users a mechanism to be alerted about the issues proactively. This in turn increased trust and adoption by the users, and the data warehouse team regained credibility and respect of the users. This also laid the foundation for future projects to address the problems impacting data quality at the source and improving reliability of the BI system.
This solution was the first step in an iteration of projects that ultimately improved the quality of data as well as other data integration problems for the business intelligence and analytics environment.
The benefits to the client included:
Significant increase in trust and adoption of the reporting and analytics environment
The added confidence allowed for follow-up projects that enabled a road map towards major improvements in the business intelligence and the corporate performance management environments
Visibility into causation of issues, which in itself improved data quality at the client
Decreased support costs and time spent researching issues for end users