ChatGPT is a great tool for synthesizing data, identifying patterns, and surfacing hidden insights. It can also create fantastic visualizations of the data. ChatGPT leverages Python libraries to analyze and visualize your data.
How to use data analysis in ChatGPT
- Start by uploading a supported file type
- Comma separated values (.csv)
- Write your prompt using the DIRECT framework in the “Ask Anything” box.
Prompting Best Practices
- Follow the DIRECT framework to make your prompt as specific as possible for your desired output.
- Good Data Analysis questions include:
- “Here’s a CSV with 6 months of sales data. Can you summarize the main trends and highlight any outliers?”
- “Analyze whether there’s a relationship between marketing spend and revenue. Create a short explanation and show it visually.”
- “This data has missing values and inconsistent date formats. Can you clean it and explain what steps you took?”
- “Using the insights from this report, can you write a short summary explaining what the data suggests about customer behavior?”
- “Given the past 12 months of user growth data, can you project likely growth for the next 3 months and explain your reasoning?”
- When preparing spreadsheets to be uploaded:
- Include descriptive column headers in the first row
- Use plain language for column headers, avoiding acronyms and jargon
- Include multiple sections and tables in a single spreadsheet
- Include empty rows or columns
- Include images which contain critical information
Technical Details
Visualization capabilities
Bar, pie, scatter, line chart, histograms, scatter plot, box plots (box-and-whisker plots), heat maps, area charts, radar charts, treemaps, bubble charts, and waterfall charts.
Number of Files
- [Normal Chat] Up to 10 files
- [CustomGPTs] Up to 20 files can be attached to a GPT as Knowledge (ChatGPT can interact with these files if the Code Interpreter capability is enabled at the GPT level)
