Skip to main content

Editing your schema

Rename columns, change types, reorganize structure, and export - manage your dataset schema without re-uploading.

Your dataset's schema - the column names, types, and structure - defines how Genuics interprets and displays your data. You can edit the schema at any time without re-uploading, making it easy to correct mistakes, improve clarity, or adjust as your analysis needs evolve.

Rename columns

Column names from raw uploads are often cryptic - "q4_nps_score", "col_12", or "Unnamed: 0". Renaming them makes your dashboards, reports, and filters much easier to read.

  1. Open the dataset and switch to the Schema tab.
  2. Click the column name you want to rename. It becomes an editable text field.
  3. Type the new name. Use something human-readable - "NPS Score" instead of "q4_nps_score", or "Customer Region" instead of "col_12".
  4. Press Enter or click away to save.

Renaming a column updates it everywhere in Genuics - any widget, report, or filter that references this column automatically uses the new name. You don't need to rebuild anything.

Change column types

If Genuics auto-inferred the wrong type during upload - or if your analysis needs have changed - you can switch a column's type at any time.

  1. Open the dataset and go to the Schema tab.
  2. Click the type dropdown next to the column you want to change.
  3. Select the new type from the list.
  4. Click Save.

Common type changes:

FromToWhy
numbernpsUnlock NPS scoring (Promoter, Passive, Detractor breakdowns)
numbercsatUnlock CSAT percentage calculations
stringcategoricalEnable dropdown filters and better grouping in charts
stringopen_endEnable AI theme and sentiment analysis
numberIDPrevent meaningless aggregation of identifier columns

See Field types and mapping for the full list of types and what each one does.

Reorganize column order

The order of columns in the Schema tab determines how they appear in column pickers throughout Genuics - widget configuration, report builders, and filter selectors. Putting the most important columns first saves time.

  1. Go to the Schema tab.
  2. Drag a column row by the handle on the left side to move it up or down.
  3. Drop it in the desired position. The new order saves automatically.

Export your dataset

You can export the current state of any dataset to a CSV file, including all schema changes, renames, and cleaning you've applied.

  1. Open the dataset.
  2. Click Export in the toolbar.
  3. Select CSV format.
  4. The file downloads with your renamed columns, corrected types, and cleaned data.

Exports are useful for:

  • Sharing a cleaned version of the data with someone outside Genuics
  • Creating a backup before making major schema changes
  • Feeding the data into another tool with your improvements applied

When to edit vs. re-upload

Editing the schema is the right choice when you need to:

  • Rename columns for clarity
  • Fix a few incorrect types
  • Reorder columns for convenience
  • Remove columns you don't need (via Data cleaning)

Re-uploading is better when:

  • The source data has fundamentally changed - new columns added, large sections removed, or a completely different structure
  • You need to replace all the data - for example, a corrected export from your source system
  • The original upload had errors - corrupted rows, encoding issues, or a truncated file

Schema changes and existing dashboards

When you edit your schema, here's how it affects things that already exist:

  • Renamed columns - widgets and reports update automatically to use the new name. No breakage.
  • Changed types - widgets that use the column in a way that's incompatible with the new type may need adjustment. For example, a chart that averages a column will break if you change that column from "number" to "categorical."
  • Deleted columns - any widget, filter, or report field that references the deleted column will show an error. You'll need to update those manually.

Next steps

If you haven't already, read Field types and mapping to understand how each type affects your analysis. For value-level cleanup, see Data cleaning.

Was this helpful?

Related articles