Clean one file. Reuse the recipe on the next.

1. Download the samples

Each file has 21 data rows and the same eight columns. They include whitespace, uppercase emails, repeated full rows, an empty row, mixed date formats, and values that need human review.

Import the original CSV files directly. Saving them in another spreadsheet application first may change the intended test values.

2. Run the test

Use the same browser and account throughout so you can access your saved recipe.

1. Import September

   Sign in to the workspace, choose Open file, and import 01-fictional-customers-september.csv. Confirm there are 21 data rows and eight columns.

2. Preview five cleanup operations

   Open Review cleaning plan. Enable trimming whitespace, removing empty rows, removing duplicate full rows, lowercasing emails, and normalizing safe dates to YYYY-MM-DD. Leave other operations off for this first test. Inspect the proposed before/after values and removed rows, then apply the plan.

3. Check the result and try undo

   The expected result is 18 data rows: two repeated full rows and one empty row are removed. Both people named Alex River should remain because their IDs and emails differ. Undo once to return to 21 rows, then redo to restore the cleaned result.

4. Save a reusable recipe

   With the cleaned result active, open Recipes → Save current. Name it Monthly customer cleanup and choose Save reusable recipe. Save it before making row-specific manual corrections.

5. Run it on October

   Import 02-fictional-customers-october.csv as a new table, without appending it to September. In Recipes, select your saved recipe and choose Run on current file. If field mapping is requested, match the same column names and inspect any compatibility warning. The expected result is again 18 rows, including both people named Jordan Glen.

6. Review remaining issues and export

   Inspect the values listed below, then try exporting the October result as CSV. If export requests review, follow the review controls without inventing missing facts. If you get stuck, tell us which step and message you encountered.

3. Check what still needs review

The five cleanup operations do not fill missing values, infer uncertain dates, or decide whether an unusual amount is correct.

The zero amounts at C09016 and C10016 are intentional and should remain zero. Company names containing a comma should stay in one cell, and Chinese names should remain readable.

Country variants such as US, USA, and United States are intentional. You can optionally try explicit field mappings after completing the main test.

4. Share a short reply

Reply in the Reddit thread where you found this kit:

1. Could you tell which values would change and which rows would be removed before applying?
2. Did saving the recipe and running it on the second file work? Where did you get stuck?
3. What CSV or Excel exports do you normally clean, and would this workflow help?

Screenshots are optional. Include the step number and error message if something failed, and keep private data and account details out of your reply.