Everything, everywhere, all at once
By 2030, many companies will be approaching “data ubiquity.” Not only will employees have the latest data at their fingertips, as we highlighted in “The data-driven enterprise of 2025,” but data will also be embedded in systems, processes, channels, interactions, and decision points that drive automated actions (with sufficient human oversight).
Quantum-sensing technologies, for example, will generate more precise, real-time data on the performance of products from cars to medical devices, which applied-AI capabilities will be able to analyze to then recommend and make targeted software updates. Gen AI agents informed by detailed historical customer data will interact with digital twins of those same customers to test personalized products, services, and offers before they are rolled out to the real world. Clusters of large language models (LLMs) working together will analyze individual health data to derive, develop, and deploy personalized medicines.
Some companies are already embracing this vision, but in many organizations, few people understand what data they really need to make better decisions or understand the capabilities of data to enable better outcomes.
Essential actions for data leaders
Enabling these visions of advanced technologies requires the data leader to activate the organization so it thinks and acts “data and AI first” when making any decision. That means making data easy to use (by creating standards and tools for users and systems to easily access the right data), easy to track (by providing transparency into models so users can check answers and automated outcomes), and easy to trust (by protecting data with advanced cyber measures and continually testing it to maintain high accuracy).
Data leaders will need to adopt an “everything, everywhere, all at once” mindset to ensure that data across the enterprise can be appropriately shared and used. That includes, for example, clearly defining and communicating data structures (that is, data hierarchies and fields) so teams understand the standards needed for a given data set and establishing clear business rules (such as naming conventions or types of data that are acceptable to collect), which will need to be revisited frequently as models, regulations, and business goals evolve.