With the growing excitement around how AI can supercharge exploratory data analysis (EDA), I find myself equally intrigued by AI's role as a summarizer and storyteller in repetitive data analysis (RDA).
The hype around how AI can enhance EDA is understandable. AI can assist end-to-end—helping users understand available datasets, write SQL queries and execute transformations, and then build meaningful visualizations. This enables an analyst or even a business user to rapidly explore and analyze their data in ways previously unimaginable.
However, organizations invest a disproportionate amount of time in understanding and refining their key metrics, continuously monitoring how different drivers impact those metrics. This is especially true as businesses experiment with new tactics, run campaigns, or test new strategies. The questions always come back to: What is working, where, and why?
Fortunately, much of this analysis follows repetitive patterns—whether it’s woven into business reviews on a weekly or monthly cadence, financial closing analysis and reports, or preparation for quarterly board meetings. These processes are ripe for productization and automation.
In this context, a higher-order framework like metric trees, which explicitly capture the underlying business processes and models, becomes a game changer. In this world, the necessary calculations and algorithms can be executed in a deterministic layer (e.g., Trace!), and AI can scan and identify patterns within those results, summarizing and telling the story of what’s happening and why.
While I’m optimistic about AI’s potential to supercharge analyst and developer-like workflows, I’m also bullish about its ability to extract narratives that bring business teams closer to the data, making decision-making faster and more effective.
