As enterprises become increasingly data-driven, there is a growing realization that true transformation cannot be confined to technical teams alone.
To unlock the full value of their data, organizations are redefining what data literacy looks like – expanding it beyond IT departments to empower non-technical employees. From marketing to HR, more professionals are being trained to read, interpret and act on data in real time.
Rahul Rastogi has a unique perspective on what modern data literacy looks like today, including how companies can close the skills gap and why data democratization is essential to thrive in complex, data-intensive environments.
At SingleStore, Rastogi leads transformative initiatives that drive innovation and growth through data and analytics. He brings over two decades of experience from Apple, where he built and managed global business strategies and delivered impactful data-driven solutions at scale.
Digital Journal spoke with Rastogi to gain an insight into the important transformation modalities for businesses.
Digital Journal: What should “data literacy” mean in 2025 for non-technical people? What should that knowledge include?
Rahul Rastogi: At this point, it doesn’t really matter how much the outlook on data literacy has changed – our lives and workflows are becoming more technical with AI’s influence growing and tools like ChatGPT making advanced tech accessible to everyone. This shift towards data democratization, where insights aren’t only confined to technical teams, means we need to meet people where they are and ensure they’re properly trained. Whether you’re technical or not, it’s better to get on top of understanding your data now because it’s only going to play a bigger role in how we work and make decisions moving forward.
Knowledge about data can include concepts like data cleansing, transformation, organization and integration but these are more technical concepts which non-technical folks don’t need to understand. They need to understand the source of data, meaning of data, conformed dimensions (like product hierarchy, customer, channel or GTM, etc.), good understanding of reference data, definitions of measures and metrics and understanding of business processes. If they understand these things well, they will be more ready to use data in their day-to-day operations to drive decisions.
Importance of data freshness, completeness of data and understanding of data quality in terms of business metrics is also important. For example, it isn’t sufficient for non-technical users to know 100 records moved from source and came to the analytics environment. Checks like total number of units sold or total value of orders in relation to what it was yesterday or same day last week would give them an indication of completeness and accuracy of data. Non-technical users understand their business well, but how they are using data to get similar insights would be key.
DJ: How can organizations make their employees more data literate besides offering data literacy training?
Rastogi: Organizations can integrate data literacy into their everyday workflows by creating cross-functional projects that analyze and interpret data. For example, customer service teams can work with analysts to review chatbot data, identifying trends to improve customer experiences. Marketing teams can partner with AI specialists to test AI-generated campaign messaging and analyze performance metrics.
In addition to data literacy training, it is imperative that organizations take a moment to catalogue data and data models such as: catalogue dimensions, hierarchies, reference data, measures, metrics, access and authorization details, ownership & stewardship details and relationships between them. The details for the aforementioned should be in business language with examples, context and with business definitions. The catalogue can be organized by business function or business process, or by GoToMarket. For example, if an organization has data that supports their financial forecasting or financial close, the data that enables this function can be catalogued.
In the world of GenAI, making data discoverable via chatbots or search is easy and can help users discover data easily. They can discover meta data, data models, sample data and even use GenAI to get confidence on data quality, completeness and accuracy.
DJ: Which organizational leaders should be involved in helping to make the workforce more data literate?
Rastogi: CIOs, CDOs and department heads should lead the effort and be responsible for creating a culture where data skills become part of everyday work. While these leaders will drive the initiatives, every leader must have a basic understanding of data’s role in decision-making, regardless of their function. From finance to marketing, leaders need to know enough about data to ask the right questions and guide their teams. Also, IT, HR and other business units can facilitate these cross-functional projects.
Data literacy is also an outcome of data democratization. It should be easy for everyone to gain access to corporate and business data, understand the data and use database tools to incorporate them in their day to day processes. Data in a fortress is less likely to be used; removing friction for non-technical users, empowering data consumption through self service tools and fundamentally changing the culture to reward use of consistent data for decision making across all levels of the organization will go a long way.
DJ: What challenges are getting in the way? What should be done about them?
Rastogi: A big challenge is the persistent use of bad or low-quality data. The root causes of this are the data feeding AI systems and LLMs, including human data entry errors, incomplete or outdated data, and issues arising from integrating multiple data sources. Bad data creates barriers and undermines confidence in data-driven decisions.
Organizations should standardize their data practices, implement regular data quality assessments and audits, and enforce standardized formats and collection methods. Prioritize data quality and invest in proper data management practices, ensuring that their initiatives are built on a solid foundation.
In organizations, data silos, lack of data integration, lack of documentation and cataloging and tribal knowledge about the data is also a big challenge. GenAI can solve several of these problems but it requires focus on governance, establishing cross functional data councils of experts who are responsible for data definitions and the clear definition of data owners. It also requires well defined data access and authorization processes, strong communication processes, laser focus on data quality measures and, most importantly, thinking about how organizations will use data to drive decisions upfront in designing data capture systems.
