Data is the essence of a winning digital transformation program and organizations leveraging data analytics can predict, build, and drive successful business outcomes. But developing the right analytic use case for a business problem can be overwhelming if the data infrastructure, platform, technologies, people, and processes are not organized, pre-defined, and acknowledged/applied.

On top of that, valuable time is invested in securing budgets, building business cases, and getting approvals from the management. The use case development opens the Pandora’s box which unveils many little-known pitfalls such as lack of documentation, unknown data sources, untrustworthy calculation rules, and data owners, among others. In many such cases, the data analysts and scientists are at the mercy of historical pieces of information maintained (or not) in Excel sheets, leading to bottlenecks in development.

When structured and organized data management practices are lacking, data analysts often encounter various obstacles that require them to make assumptions to find a solution. This can result in data inconsistencies and noticeable issues with data quality. In scenarios like this, a familiar question during the review with the business team could be, “Why is 30% of sales contributed by the ‘null’ category?”

Industrialization of analytics use cases without proper data foundations produces substandard outcomes and fails to deliver the promised impact. Ineffective solutions also question the skills and credibility while damaging the trust in the people building and managing data for the business users.

To overcome such failures, unlocking data discoverability, accessibility, observability, security, literacy, and quality are vital for organizations. Data governance can not only address the issues but also ensure all activities follow governance guardrails to avoid cascading failures due to legacy ecosystems.

Most organizations have moved or are in the process of moving to cloud-based analytics solutions. Hence, organizations should focus on data lake optimization as a bare minimum objective for fully leveraging data governance. Many data lakes overlook the significance of data quality in the raw zone and perform basic data validation checks only when data is moved to the curated zone for analytics.

This is one of the main reasons which lead to significant challenges for organizations. Thus, it is crucial to establish a well-defined data ingestion process that includes basic governance elements. To achieve this, a data catalog is an essential tool for capturing the following 6 types of information:

• Source system IT owner
• Metadata scanning of the system
• Tagging critical and PII information
• Data Owner and Steward identification
• CDEs (critical data elements) identification
• Data quality set up for CDEs and remediation process

By establishing a standard checklist for data ingestion, data will find a home in the data lake with enhanced credibility and quality. A cohesive data catalog will make information readily available and trustable, shortening development cycles and producing desirable outcomes.

In many organizations, data literacy and the risks associated with it remain at large when a plethora of information is undocumented. Such crucial institutional knowledge is bound to erode over time as employees and contractors move on and spreadsheets grow stale. By executing robust governance principles and adhering to them consistently, organizations can unlock the full potential of their analytics investments.

With well-managed data assets and streamlined analytics solutions, users and data professionals will be empowered to find, access, and utilize data more effectively, leading to faster implementation and increased value for the organization.

About the Author

Ashish Arora is a Group Head of Data and Analytics at Central Group where he leads BI, analytics, and data governance for its retail business. He has over 18 years of experience leading different strategic analytics, transformation, and data governance programs.

Arora is outcome-driven and passionate about solving critical business challenges using data and analytics solutions. He has extensive experience establishing D&A teams, setting up data governance, and uplifting Analytics maturity in different organizations.

His professional experience also includes Global Head of Data Strategy at British American Tobacco and Head of Data and Analytics at AirAsia.