5 Essential Components of a Data Strategy
Building a comprehensive data strategy for your organization can be a daunting task. Where do you start, and how do you put all the proper pieces in place for a suitable strategy to bring value to your company from data? The key is to ensure that the strategy focuses on the strengths and needs of the individual company.
1. Identify and Describe: A data strategy is “a roadmap and plan to identify what to do with a company’s data and to support accessing, sharing and managing the content,” says Evan Levy. In identifying and referencing content, look for subject area and data element names/descriptions, content type (structured, unstructured, semi-structured) and metadata. This component also includes the tools and methods to enable and automate the collection, publishing and managing content details as well as the business and technical-level details.
2. Provision and Share: The key to providing and sharing data between systems is to determine packaging (files, transactions, data streams, etc.), content and formatting (values, formats, etc.) and package metadata details (location, origin, availability, changes, etc.). This component also defines methods to allow data availability, such as support for internally and externally delivered content, production support and change control (errors, fixes, versions, etc.) and interface and access to allow data delivery.
3. Stage and Store: Areas to consider for storing and sharing enterprise content include master/reference data (systems of reference), business event details (transactional history), external reference and descriptive content and processing (applications, reporting and data integration). Also decide the tools and technologies to be used for content storage, specifically the storage system (DBMS, flat files, cloud, Hadoop, etc.) and the access method (API, Web services, applications, etc.).
4. Integrate and Move: It is necessary to consider the movement and transformation of source data for use by downstream systems, including identification and matching; cleansing, standardization and acceptance of content; and metadata and data lineage details. Determine the data movement/migration infrastructure to be used for bulk data movement and processing and application/transaction messaging (ESB).
5. Govern and Manage: What are the company’s policies for managing the data? Data governance methods and processes should cover information access policies, and methods and process for supporting data access and resolving conflict. Data management methods and practices should include the adoption and usage of data standards and the tactical deployment of data policies into applications and data usage.
Evan Levy, vice president of business consulting at SAS, shares the five necessary components for building a data strategy, based on his new course at TDWI.