Uploading CSV and Excel files

Uploading a file is the fastest way to start. ChartBase accepts .csv and .xlsx.

What happens on upload

1. The file is stored, and the first rows are read to detect each column's type.
2. The rows are converted into a columnar format built for analytical queries.
3. Charts read that copy, never the original file.

That last point matters: the file you uploaded is not what your charts query. If you clean the data afterwards, the cleaned version is what gets queried.

Check the detected types

Type detection is good but not psychic. It is worth a look before you build anything, because a column's type decides what you can do with it:

• A date read as text cannot be grouped by month or used for time intelligence.
• A number read as text cannot be summed.

You can correct a column's type from the dataset's settings, and doing it now costs seconds where doing it later means revisiting every chart built on it.

First row as headers

The first row is treated as column names. If your file starts with a title row or a blank line, tidy that before uploading — an ordinary spreadsheet with data starting in row 1 works with no configuration at all.

When a file is the wrong tool

A file is a snapshot. If the underlying data changes and you need the dashboard to follow it, connect a database instead — the same modelling and charting work either way, but the data can be refreshed on a schedule.