AI For Data Analysis

AI has collapsed the distance between a business question and a defensible answer. Work that used to require an analyst, a week, and a pivot-table specialist (cleaning messy exports, exploring a dataset, charting a trend, drafting the findings) can now be done conversationally by the person who owns the question.

This workshop teaches that workflow end to end, using the data-analysis capabilities of modern AI assistants against real, messy, spreadsheet-shaped data. You will clean and reshape exports, run exploratory analysis, produce charts a decision-maker can read, and extract trends and simple forecasts, all without writing code.

Half the curriculum is about trust, because numbers are where AI errors are most expensive and least visible. You will build the verification habits professional analysis requires: tracing figures to source, sampling, reconciliation checks, and knowing which questions an AI answers reliably and which it does not. The capstone is an end-to-end analysis of a real dataset, delivered as a one-page decision memo.

Participants will learn to:

• Frame a business question so an AI assistant can analyze it usefully
• Clean and reshape messy spreadsheet data with AI assistance: duplicates, formats, joins, and gaps
• Run exploratory analysis conversationally: distributions, segments, outliers, and comparisons
• Produce decision-grade charts and choose the right visualization for the claim
• Extract trends and simple forecasts, and state their limits honestly
• Verify AI-produced numbers: source tracing, sampling, and reconciliation checks
• Package findings as a one-page decision memo a leader can act on