If the long-term success of any data management undertaking is to be ensured, the focus must be on the near-term success of the current initiative. And, simultaneously, the organization must never lose sight of its long-term strategic goals.
Gone (or should be) are the days of the “one-and-done” pilots and “throw away” proofs of concept that show some initial promise but never materialize into sustainable value for the organization. If companies are going to make the shift successfully to enterprise-wide, longer-term, more strategic projects, they need to consider solutions and tools that enable the broader data management vision. These also will help the organization capitalize on the immense value that data can provide to the business.
Data has to be discovered, acquired, organized, filtered, and cleansed before it offers real value to the end consumer. The data discovery process normally begins with the framing, socialization, and agreement of a business use case for the data project. By developing a set of prioritized use cases, the organization can clarify the functional areas that will be impacted and establish both strategic guidance for the initiative and a set of success criteria to aid measurement.
A clear understanding of what the organization hopes to accomplish and how it will measure success is important so that the right data sources can be leveraged for the acquisition of the necessary data.
Making the Right Data Decisions
Now the organization is ready to acquire, store, and integrate the data that it will use to help it reach the goals and objectives it identified prior to (and during) data discovery. A host of technology solutions are available today for acquiring all types of data (internal, external, structured, and unstructured), and equally as many tools are available for storing this data (from traditional enterprise data warehouses to Hadoop and cloud-based data stores). Often, the size and complexity of the data and how the organization hopes to use it will influence the overall storage strategy and selection of tools.
It could be argued that storage is not nearly as important a decision as the one(s) concerning how best to integrate this data with existing data and legacy data solutions. Failure to create an integrated and optimized data landscape can leave the organization unable to drive actionable value from its legacy technology investments and recent technology acquisitions.
If the organization is unsuccessful in the way it manages any of these vital steps, this will diminish the value that the data offers, and the resulting insights may be questionable at best. Companies should, therefore, view data management as an enterprise program that is ongoing – a program that leverages solutions and tools that are integrated and optimized to provide one cogent outcome while ensuring that they can scale for enterprise-wide adoption.
A Journey, Not a Destination
Equally important is the concept that information management is ongoing. It should build on processes that are interrelated to form a cohesive, comprehensive, and well-defined strategy (rather than being done in a silo – i.e., no one-and-done initiatives and no throw-away efforts). It is important to focus on the processes behind how one goes about any data management or information management initiative. The desire to utilize the tools immediately without first understanding the available data and devising a proper strategy that allows the organization to truly benefit from it is a recipe for disaster. Having a robust and agreed-upon data governance strategy in place is paramount to the success of any data initiative.
Data governance is the process by which standards and requirements are created and agreed upon for the collection, identification, storage, and use of data. In the era of big data, data governance must include structured, semi-structured, and unstructured data, registries, taxonomies, and ontologies, as it contributes heavily to organizational success through the application of repeatable and compliant practices.
Other key areas that must be considered for the success of an enterprise-wide data management program include data security and information life cycle management (ILM), metadata (metatags), master data, model management, etc. These are critical aspects of a successful data and information management program.
Building the Right Foundations for Success
The path to achieving a first win is important to the longevity and expansion of the program. Big data projects often flounder if the right foundation for further exploitation is not created from the beginning. Frequently, the wrong foundation involves incorrect incubation, or incubation in a manner in which a concept cannot be converted into a full-scale outcome because the people, technical, or process considerations of a full-scale implementation were not incorporated into the original project.
When ecosystem considerations are abstracted from the original project, the leveragability of the project is severely limited. Therefore, it is always recommended to build into the first project all ecosystem considerations, including data governance, ontology, classification, metadata, data cleansing and quality, master data management, and scalability (both up and out). This allows for hypotheses, which have been tested in a limited context, to be easily translated into business process improvements and results for the current and future outcomes. This forward-looking approach is often the mark of a successful initial implementation.
The same basic “blocking and tackling” continues to hold true even with the big data paradigm: focus on platforms, processes, and programs that guide today’s data-driven organizations as they expand their sights on the acquisition and use of new data and the achievement of enterprise-value outcomes. This focus will allow for the proper enterprise-wide scale-out that is needed to achieve new heights of operational efficiency and organization profitability.
Scott Schlesinger is a Principal within Ernst & Young LLP’s National IT Advisory Practice, and serves as EY’s Americas Leader for Business Intelligence and Information Management.
The views expressed herein are those of the author and do not necessarily reflect the views of Ernst & Young LLP.
Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise.