How to Make a Pie Chart from a CSV File

Nearly every tool that holds data can export a CSV: analytics dashboards, CRMs, banks, survey platforms, accounting software. That makes CSV the universal bridge into a chart. Here is how to cross it, including the messy parts real CSV files bring along.

What the file needs to contain

A pie chart needs exactly two things per slice: a name and a number. So your CSV needs at least two usable columns, something like:

category,amount
Housing,1450
Food,620
Transport,410
Utilities,230

A header row is optional. The parser looks at the first row, and if it reads as text while the rows below read as numbers, it treats it as a header and skips it. Extra columns are fine, and so are files with thousands of rows, though a chart wants far fewer slices than that, which we will get to.

Uploading and mapping

Drag the file onto the upload area in the data panel, or click it and browse. Parsing happens instantly and locally: the file is read by your browser and never uploaded anywhere, which matters when the CSV is a customer export or your bank statement.

If the file has more than two columns, the editor asks which column is labels and which is values. Pick them and the data table fills in, with the live chart drawing beside it.

Reading the warnings

Real CSVs are rarely clean, so the parser reports what it did instead of failing silently. After import you may see warnings like “3 rows skipped: no numeric value”. Common causes and what to do:

• Currency and formatting. Values like “$1,450.00” or “1.450,00” are cleaned automatically. You should not see warnings for these.
• Empty rows and totals. A blank line or a “Total” row at the bottom gets skipped or, worse, charted. Delete a lingering total row in the data table, otherwise your chart shows every category at half its true share.
• Text in the value column. “N/A”, “pending”, or a stray note becomes a skipped row. Type the correct number into the table if the row matters.
• Duplicated categories. Two rows named “Food” become two slices. Merge them in your source file, or combine the values manually in the table.

The one-minute check after any import: does the number of slices match the number of categories you expected, and does the biggest slice make sense?

From 200 rows to 6 slices

CSV exports are usually itemized: every transaction, every order, every response. A pie chart wants aggregates. If your file has one row per transaction rather than one row per category, aggregate first in your source tool (a pivot table, or a GROUP BY, or the report’s own summary view), then export.

If the file is already aggregated but has a long tail, say 30 product categories where the bottom 20 are each under 2 percent, use the group-small-slices button. Set a threshold and everything under it collapses into one Other slice. The chart keeps its accuracy and becomes readable.

Finishing the chart

From here the CSV origin no longer matters. Turn on percentage labels, which are rounded to always sum to 100. Add the category names to slices if the chart will stand alone. Give it a title that names the data and period: “Support tickets by category, June 2026”. Then export as PNG for documents, SVG for design tools, or PDF for print, or copy the chart straight to your clipboard.

One last habit worth keeping: save the chart in the My Charts list before you close the tab. Next month, when the same tool emits the same CSV with new numbers, you can duplicate the saved chart, re-import, and keep every styling decision you made today.