Column settings
Types, display formats, default aggregation, sorting, folders, and hiding columns.
Each column carries settings that travel with the dataset, so every chart built on it inherits them instead of being formatted one at a time.
Data type
Decides what a column can do — grouped by month, summed, compared. Correct a wrong detection here rather than working around it in each chart.
Display format
How values are shown: decimal places, thousands separators, currency, percentages, and date formats. Set once, applied everywhere the column appears.
Default summarization
What a numeric column does when dropped into a value well — sum, average, count and so on. If a column is an identifier rather than a quantity, this is where you stop it being summed into a meaningless total.
Sort by another column
The classic case: month names sort alphabetically (April, August, December…) unless you tell them to sort by a month number. That is what this setting is for.
Display folders
Group related fields under a heading in the field list. Slash-separated names nest one level, which keeps a wide table navigable.
Hiding a column
Removes a column from the field pickers so nobody builds with it by accident.
Hiding is presentation, not permission. The column is still in the dataset, still queryable, and any chart already using it keeps working. Anything that genuinely must not be seen has to not be in the dataset — or be restricted with row-level security.