At Data Summit's opening keynotes, John Lewis, CKO, Explanation Age LLC, Shiva Pullepu, vice president, AI and industry solutions, CrateDB, and Sami Akbay, VP, product management, insightsoftware, each examined the many new paradigms shaping the world of data, from data storytelling to real-time data unification and the way generative AI (GenAI) is transforming business intelligence (BI).
The annual Data Summit conference returned to Boston, May 14-15, 2025, with pre-conference workshops on May 13.
Kicking the keynotes off, Lewis delved into the way the AI Revolution is requiring us to adopt a different model of work—storytelling. Rivaling the measurement of work as a process during the Industrial Revolution, today’s AI-defined landscape necessitates organizations to examine data as a “storytelling” process, where data visualization is driven by stories.
Lewis contextualized his presentation with the fact that, according to the studies of neuroscience, humans are wired for story. Story is a communication strategy, translating your entrepreneurial journey into a consumable story. This captures details about insights, data, and lessons learned for others to benefit from.
Lewis examined different patterns of stories, ranging from linear paths to more cyclical ones. In the latter pattern, energy pushes transformation, bringing us continuously around ideation, execution, and reflection as challenges arise.
“At work, that’s really what we’re doing—we’re telling stories. But we don’t realize we’re in the story when we’re doing it,” said Lewis. This means that storytelling is not just a communication strategy—it’s an operational strategy.
The nature of storytelling enables us to measure the status of our initiatives, evaluating progress against where we are in the story. Are we in the ideation phase, the disruption phase, or the investigation phase? These examinations help us to not only tell the story, but guide us through it as we exist within them.
When enterprises are experiencing challenges, the story model helps pinpoint where that obstacle lives—and how to solve it. Lewis suggested that organizations should identify their weakest capability within the story framework—whether its ideation, expectation, affirmation, automation, disruption, or investigation—and identify what data and visualizations will strengthen that weak point.
During Pullepu’s presentation, he explained that creating a foundation for successful AI deployment is vital. With enterprises in a continuous race to implement and see tangible ROI from AI projects, many are shocked to discover that their data infrastructures are incredibly fragmented, preventing that success they seek.
“Today’s customers want real-time insights…so we cannot deny the fact that intelligence is crucial in building any application today,” said Pullepu. “Data is at the front and center of driving [intelligence].” Yet, with data relegated and trapped in siloes, intelligence becomes a distant dream rather than a tangible reality. Intelligent apps without live, unified data is like a security system watching old tapes during a crime, explained Pullepu.
The technical challenges creating this state of data are rampant, from an increase in data source diversity to continuously growing data volumes and highly disruptive trends such as AI and large language models (LLMs).
The solution is a real-time unified data layer, according to Pullepu, capable of reducing the time it takes to access data, as well as decreasing time-to-insights and time-to-value. With CrateDB, enterprises can implement a streamlined architecture that best supports AI initiatives, defined by a unified, real-time data foundation.
Centralizing queries and aggregations, full-text search, vector search, and AI integrations within a single database, CrateDB is set apart by its widespread application across different modalities, from SQL to NoSQL, search, and vector. Many CrateDB customers have achieved significant cost savings, operational efficiency boons, and easy scalability, noted Pullepu.
“Working with data can be better,” said Pullepu. “CrateDB is one of those unique solutions that makes your life easy.”
During the last presentation, Akbay explained that while AI is drastically changing every facet of business, GenAI did not “kill” BI. Instead, AI has transformed it, moving from passive, static dashboards to proactive, conversational insights. As this infrastructure changes, enterprises will need to adapt, embedding intelligence into workflows, agents, and applications to invite more advanced analytics.
We’re witnessing the evolution of BI into something far more powerful, said Akbay, propelled by the fact that “everyone is trying to become data-driven.” Various new paradigms—such as conversational insights, democratized data access, self-service at all skill levels, and proactive intelligence—are shaping a new era of BI.
Akbay emphasized the fact that with conversational experiences becoming the new norm in data, enterprises will need to meet users where they are. Old ways of BI, defined by separated, fragmented tooling and learning curves, are no longer valid. The way forward is shaped by embedding analytics in existing workflows and familiar applications, acting as an invisible assistant operating with seamless integrations.
Though seemingly entirely transformative, some elements of data management remain the same, explained Akbay. Themes of efficient data engineering pipelines; high-quality, storytelling data visualizations; reliable reporting infrastructures; adaptive interfaces; and robust data access and governance are still significant players in the BI realm.
“Moving forward, we need to build things that are AI-native. It is changing everything we touch, everything we do,” said Akbay. With AI, enterprises can drive compelling narratives that connect data to outcomes, actionable recommendations, and seamless delivery within existing workflows and applications.
In the AI era, Akbay argued that BI practitioners will need to embrace natural language in an increasingly conversational world. Furthermore, they must develop hybrid AI and human analytical approaches, acknowledging that GenAI is BI’s evolution—not its replacement. AI amplifies what was already working in BI, and the future belongs to those who combine the best of both.
Many Data Summit 2025 presentations are available for review at https://www.dbta.com/datasummit/2025/presentations.aspx.