Import CSV and Excel files the right way

Chartroom reads files on your device: nothing is uploaded when you import. This tutorial explains what happens to your file and how to fix the usual surprises.

Choose the file

Open Data → Import or paste data… (shortcut I) and press Choose a file…, or drag a file onto the dialog. Supported:

• CSV, TSV and TXT — comma, semicolon, tab or pipe separated. The separator is detected from the most consistent column count.
• Excel workbooks (.xlsx) — read locally. When the workbook has several sheets, an Excel sheet list appears; the preview switches when you pick another sheet. Formulas are not recalculated: Chartroom uses the values Excel stored in the file.
• JSON — an array of objects ([{"month":"2026-01","sales":120}, …]) or { "headers": […], "rows": […] }.

Old .xls files need to be saved as .xlsx or CSV first.

Check the detected columns

The preview lists one chip per column. A column becomes number or date when at least 90 % of its filled cells parse that way:

• Numbers may use thousands separators, currency signs, % and (12) for negatives. A decimal comma (1.234,5) is detected per column, so a German export and an English one can live in the same project.
• Dates can be 2026-03-04, 2026-03, 04.03.2026, 03/04/2026, Mar 2026 or 2026-Q1. When a slash date never reveals whether the day or the month comes first, Chartroom reads it day-first. Use ISO dates (2026-03-04) to be sure.
• A column of plain years (2024, 2025) counts as a number, which keeps it sortable.

If a column you expected to be numeric shows text, a few cells probably contain words such as n/a. Fix them in the source or accept that those cells are skipped in charts.

Replace or append data later

Open the dataset and use:

• Replace data… for a new export of the same table (e.g. this month’s file). The steps keep their settings and are marked outdated until you run them again.
• Append rows… to add new rows below the existing ones. Columns that are new are added; missing cells stay empty.

Before every replacement or append, Chartroom keeps the previous version as a snapshot. Under Snapshots and comparisons you can compare both versions or restore the old one.

Set the update interval

The Expected update interval of a dataset (daily, weekly, monthly …) is used by the Freshness check step and as the default for new Reminders.

Limits

A dataset holds up to 50,000 rows and 200 columns; a project keeps up to 5 MB of CSV text per dataset. Tool steps accept up to 2 MB per call — filter large tables first with Filter and Save as new dataset.