Tools and Workflows
Spreadsheets, AI, and When to Use What
Different tools suit different tasks. This chapter helps you choose and combine tools effectively.
The Tool Landscape
Spreadsheets (Excel, Google Sheets)
Best for:
- Smaller datasets (under ~100,000 rows)
- Ad-hoc analysis
- Familiar, accessible
- Calculations with formulas
- Basic charts
- Manual data manipulation
Limitations:
- Performance with large data
- Complex analysis
- Version control
- Collaboration challenges (Excel)
AI Assistants (Claude, ChatGPT, etc.)
Best for:
- Natural language questions
- Explaining concepts
- Suggesting approaches
- Pattern recognition
- Data cleaning suggestions
- Generating code
- Interpreting results
Limitations:
- Context window limits
- Can make mistakes
- Need to verify outputs
- Not connected to live data
Business Intelligence Tools (Tableau, Power BI, Looker)
Best for:
- Dashboards and visualization
- Large datasets
- Shared reporting
- Regular monitoring
- Interactive exploration
Limitations:
- Learning curve
- Cost
- Setup required
- Less flexible for ad-hoc
Programming (Python, R, SQL)
Best for:
- Complex analysis
- Large datasets
- Reproducibility
- Custom modeling
- Automation
Limitations:
- Learning curve
- Setup overhead
- Slower for simple tasks
When to Use What
Quick, One-Time Analysis
Start with: Spreadsheet or AI
Small dataset, simple question, don't need to repeat? Open Excel or paste into AI.
Regular Reporting
Use: Spreadsheet templates or BI tools
Build once, refresh regularly. Templates save time.
Exploring New Questions
Use: AI first, then spreadsheet or code
AI helps you figure out the approach. Then execute with appropriate tool.
Large Dataset
Use: BI tool or programming
Spreadsheets slow down with large data. Use tools designed for scale.
Learning New Techniques
Use: AI for explanation, then practice in tools
AI explains concepts clearly. Then apply in real tools.
Combining Tools Effectively
The AI + Spreadsheet Workflow
- Plan in AI: Ask AI how to approach the analysis
- Prepare in spreadsheet: Clean and structure data
- Analyze with both: Use AI for complex questions, spreadsheet for calculations
- Verify: Check AI outputs against spreadsheet calculations
- Communicate: Use AI to help draft summaries
Using AI with Spreadsheets
Getting help:
- "Write an Excel formula to [calculation]"
- "How do I create a pivot table for [task]"
- "What's wrong with this formula: [formula]"
Analysis:
- Paste data into AI
- Ask questions
- Request calculations
- Get interpretation
Practical Spreadsheet Tips
Essential Functions
SUMIF/SUMIFS: Sum based on criteria COUNTIF/COUNTIFS: Count based on criteria AVERAGEIF: Average based on criteria VLOOKUP/INDEX-MATCH: Look up values IF: Conditional logic TEXT/VALUE: Convert between text and numbers CONCATENATE/&: Combine text
Pivot Tables
The most powerful spreadsheet feature. Learn it well.
Use for:
- Summarizing by categories
- Cross-tabulations
- Quick calculations
- Data exploration
Data Validation
Prevent errors by restricting inputs:
- Drop-down lists
- Number ranges
- Date restrictions
Named Ranges
Name your data ranges. Makes formulas readable.
Workflow Examples
Example 1: Monthly Sales Report
- Export sales data from system (CSV)
- Open in spreadsheet
- Clean: remove duplicates, fix formats
- Create pivot table: sales by product/region
- Add month-over-month comparison
- Create charts
- Use AI to help write summary
- Share report
Example 2: Customer Segmentation
- Export customer data
- Paste sample into AI: "How should I segment these customers?"
- AI suggests RFM approach
- Calculate RFM scores in spreadsheet
- Ask AI: "How do I categorize these scores?"
- Apply categorization in spreadsheet
- Analyze segment differences
- Create recommendations
Example 3: Ad-Hoc Question
- Business question arises
- Identify needed data
- Paste into AI: "What's the relationship between X and Y?"
- AI provides analysis and interpretation
- Verify key numbers in spreadsheet
- Communicate findings
AI Prompt: Tool Selection
Help me choose the right approach for this analysis.
What I need to do: [Analysis task]
Data size: [Approximate rows/complexity]
My tools: [What you have access to]
My skills: [What you're comfortable with]
Timeline: [How quickly needed]
Repeatability: [One-time or ongoing]
Recommend:
1. Best tool/approach for this task
2. Step-by-step workflow
3. What to watch out for
4. Alternatives if primary approach doesn't work
What's Next
Templates for common analyses.
Next chapter: Analysis templates.