Use AI to clean and summarise a spreadsheet in words
Choose the right chart for a question
Interrogate an AI-generated insight sceptically
Ask in plain language
Paste a small table (or upload the sheet) and ask: 'Summarise totals by category, flag the top 3 and bottom 3, and list anything inconsistent.' Modern assistants can also write the spreadsheet formula for you - ask for the formula, learn it, reuse it.
Chart choice is argument choice
Trend over time: line. Comparison across categories: bar. Share of a whole (few parts): pie or stacked bar. Relationship between two measures: scatter. The chart you pick quietly frames the conclusion - choose it deliberately.
Sceptical interrogation
Ask the model to state its assumptions and show the calculation. Recompute one figure yourself. If it cannot show working, treat the insight as a hypothesis, not a finding.
Key terms
Data cleaning
Fixing inconsistencies, duplicates, formats.
Aggregation
Summarising rows into groups (totals, averages).
Formula generation
AI writing spreadsheet formulas for you.
Assumption audit
Asking a model to expose what it assumed.
🛠 Hands-on task: Insight memo from 20 rows
Tools: Google Sheets + chat assistant
Create/collect a 20-row table (sales, marks, airtime - anything).
Ask AI for summary + top/bottom 3 + inconsistencies.
Write a 5-line insight memo; verify one number by hand.
Key takeaways: Before you go, three takeaways from this lesson. One: Use AI to clean and summarise a spreadsheet in words. Two: Choose the right chart for a question. Three: Interrogate an AI-generated insight sceptically. And your hands-on task: insight memo from 20 rows. Create/collect a 20-row table (sales, marks, airtime - anything). When it is done, mark the lesson complete on your dashboard, keep the artifact for your portfolio, and step into the next lesson. I will see you there.
