C3 AI Documentation Home

Quick Start: Spreadsheet or CSV to Dashboard

This quick start shows the fastest path from a CSV or spreadsheet file to a running application with charts and filterable data tables.

Before you begin

Prepare the following before starting:

  • A CSV file with column headers in the first row, UTF-8 encoding, and a file size under 20 MB.
  • Access to a C3 Code workspace with the Admin role.

Step 1: Open C3 Code and describe your application

  1. Navigate to your C3 AI environment URL.
  2. Enter a prompt in the prompt input that describes the data and the visualization you want.

Example prompt:

I have a CSV of monthly sales data by region. Build me a dashboard that shows total sales by region as a bar chart, a trend line by month, and a filterable table of the raw data.

Being specific about the chart types and the fields you want to display produces better initial results. Use the exact column names from your CSV file if you know them.

Step 2: Attach your CSV file

Select the attachment icon in the prompt input area, then select your CSV file.

C3 Code reads the file, infers the data schema (column names and data types), and uses that schema along with your prompt to generate the application.

Select Generate to submit.

Step 3: Review the generated dashboard

When generation completes, the workspace opens in Preview mode. Verify the following:

  • The chart types match what you requested (bar chart, line chart, table).
  • Column labels reflect your CSV headers.
  • The filter controls appear and respond correctly.
  • Data is loading and the values look reasonable.

If the data does not load, the most common cause is a mismatch between the column names in your prompt and the actual headers in your CSV file. Use a follow-up prompt with the exact column names to correct this.

Step 4: Configure visualizations

Switch to Configure mode to adjust the charts and layout without writing code.

In Configure mode you can:

  • Change the chart type for any panel (bar, line, pie, table).
  • Select which columns map to each axis.
  • Reorder or resize panels in the layout.
  • Adjust filter fields and default values.

For changes that require code, such as custom aggregation logic or computed fields, switch to Code mode.

Troubleshoot common CSV issues

The file is not recognized or produces no data

Verify that the file is saved as a CSV with UTF-8 encoding and that column headers are in the first row. Files exported from Microsoft Excel as "CSV UTF-8 (Comma delimited)" meet these requirements.

Data types are incorrect (numbers treated as text)

Numbers stored as text in the CSV, often caused by currency symbols or thousand-separator commas, prevent numeric operations. Remove formatting from number columns in your spreadsheet tool before uploading.

The file size exceeds the limit

C3 Code accepts files up to 20 MB. If your file is larger, filter or aggregate the data in your spreadsheet tool before uploading. For large datasets, connecting to an existing C3 Type is more appropriate than uploading a file.

Charts show the wrong fields or the wrong data

The prompt may reference column names that differ from the actual headers in the file. Reprompt using the exact column names as they appear in the CSV header row.

The encoding produces garbled characters

The file is not UTF-8 encoded. Open the file in your spreadsheet tool and re-save it with UTF-8 encoding.

See also

Was this page helpful?