Eight key elements for fueling the data-driven enterprise
The term “data rich, information poor” refers to organizations that have not established the critical components needed to transform their data into actionable insights. Attempts to become data-driven are often stymied by ambiguity surrounding the project, and a lack of awareness of the people, process and technology considerations vital to the success of these initiatives.
To become data-driven, organizations must provide decision-makers with timely access to clean, consistent, reliable and actionable information, while avoiding the myriad challenges that lie in the way. With that in mind, let’s look at eight critical elements that underpin a successful, long-term information management strategy—and how organizations can implement them to become truly data-driven.
It’s essential that any enterprise information management initiative have strong C-Suite sponsorship. However, effective governance programs also must include representatives from all areas of the business. A blended team of clinical, IT, and business stakeholders is critical to establishing priorities, obtaining funding, eliminating roadblocks, and monitoring the progress of the implementation.
In order to make accurate business decisions, it’s essential that leadership have access to clean, consistent, and reliable information. Data governance is a key discipline for addressing this requirement, as it encompasses the roles, responsibilities and business processes required to ensure greater accountability for data quality, as well as the definitions and formulas for the organization’s Key Performance Indicators (KPIs). These programs have historically been led by IT but, to be successful, data governance initiatives should be spearheaded by business leaders.
Since these individuals have ultimate accountability for one or more types of enterprise-wide reference data, or master data, they are essentially “data owners” for the governance project. Because these employees are often director-level or higher, it’s important that they have significant support from subject matter experts, or “data stewards,” in their respective business area in order for the data governance initiative to run smoothly.
Master Data Management (MDM)
MDM is closely intertwined with data governance, and is a set of tools and associated methodologies that foster the integration and maintenance of master data. Data stewards must be able to easily maintain the master data domains for which they are responsible. As such, it’s best practice for companies to develop an MDM strategy in conjunction with their data governance initiative. Those that neglect to do so not only put their data governance program at a higher risk of failure, they also significantly increase the likelihood of incomplete or inaccurate information flowing through the organization.
Metadata management helps companies develop a more holistic understanding of their data by providing easy access to data definitions, formulas, and other essential details. With features that identify all database tables, reports, dashboards and other components that could be affected by a database change, metadata management solutions enable organizations to conduct impact analysis prior to making any decisions.
These solutions also facilitate data lineage, which provides a visual depiction of the data’s origin and any changes that occurred prior to it appearing on an executive dashboard or report. These and other capabilities create a high degree of data transparency, and can greatly improve users’ confidence in the final information.
Business Intelligence (BI)
BI is the most widely known information management discipline—and arguably the most critical in realizing the greatest ROI from enterprise data. BI tools enable users to drill down into subsets of information, conduct ad hoc queries, run predictive modeling, and a variety of other functionalities essential to becoming truly data-driven. While BI’s benefits are evident, many enterprises struggle in this area. A key challenge hindering more widespread BI adoption is that many organizations have a poor information management platform, also known as data architecture.
The journey from raw data into actionable analytics is a complicated process. To be successful, companies should adopt a three-tiered data architecture approach in which each layer is designed and modeled to meet specific objectives. The first component is a “landing area” into which data from disparate systems is extracted as is, followed by a “conformance layer” into which this raw data is integrated. The final component is the “analytic layer,” in which data is transformed into a usable format suitable for self-service analytics and other BI initiatives.
There are just two remaining components that underpin a successful information management strategy: data acquisition and technical architecture.
In order to populate the data architecture outlined above, companies need access to data acquisition tools, also known as Extraction/Transformation/Load, or ETL, tools. Additionally, it is vitally important to develop a comprehensive strategy to detect changes in the source system data. Many organizations fail to develop a “Change Data Capture” strategy and thus are unable to easily maintain historical data.
A robust technical architecture composed of dedicated, properly configured servers is also required to meet the growing demand for high performance and reliability.
The eight elements outlined above are key considerations for companies to make the most of their information. However, it’s incredibly common for organizations to struggle with some, or even all of these, due to challenges with the people, processes, and technologies involved.
Obtaining a better understanding of each component and how it fits into the broader information management approach is essential for companies to become truly data-driven. Given how pervasive information is in the enterprise today, the alternative simply isn’t an option.