Which way should I dedupe?
Loopster has two tools. Both delete rows.
Dedupe is a one-off clean-up. You open it for one column, see every value that appears more than once, tick the ones you want, and delete the extra rows.
Auto-dedupe rows is a table setting. The app says: "Delete duplicate rows with the same value in one column before enrichments run."
Use Dedupe when you want to see the duplicates and pick which ones to delete. Use Auto-dedupe rows when you want the table to act on a repeated value by itself, with no list to review.
How do I remove duplicates once?
Open a column's header menu and click Dedupe. The window opens straight to Dedupe rows by <column>.
Read the list. Each value that appears more than once has a count, for example "2 rows".
Tick the values you want to clean up. The header tick box selects every duplicated value.
Click the red button, for example Delete 1 duplicate rows.
The rows are deleted at once. There is no second confirmation and no preview of the rows. A message says how many duplicate rows were deleted.
You can also click More table options, then Dedupe. That opens Select a column to dedupe first. Open the Column menu and pick a column. The window changes at once to the same list, with an extra Back button.
The window says: "Review duplicates found in the <column> column." and "Every row in this table was checked, filters and search included. Values must match exactly — capitals and spaces count. Only the first occurrence of each value is kept."
What counts as a duplicate?
Two values match when they are exactly the same. The window says "capitals and spaces count".
What this means:
"alpha" and "Alpha" count as different values.
The first row in row order is kept. The later ones are deleted.
Blank values are not listed, and the blank rows stay.
How do I switch on Auto-dedupe?
Click Table settings next to the table name.
Click Deduplication to open it. It is closed until you open it.
Switch on Auto-dedupe rows.
Next to Dedupe via column (marked with *), pick the column to compare.
Under When duplicate rows are added..., choose one:
Keep oldest row: "Rows that match existing rows will be deleted." This is the default.
Keep newest row: "Existing matching rows will be deleted; enrichments will run on the new row."
The app also says: "Blank values and values longer than 200 characters are ignored."
What does Auto-dedupe do when I type a duplicate value?
When you type a value into a new row and the column already has that value:
With Keep oldest row, a message says "Duplicate value was not saved." and the new row disappears.
With Keep newest row, the older row is deleted and the new row stays.
Adding blank rows removes nothing, because blank values are ignored.
Troubleshooting
My duplicates did not disappear when I switched on Auto-dedupe
What you see: You switch Auto-dedupe rows on and pick a column. Duplicates that were already in the table are still there.
Why it happens: The options say "When duplicate rows are added...". Auto-dedupe rows works on values you type from then on.
What to do: Use Dedupe to clean up the duplicates already in the table. It lists them and deletes the ones you tick.
I deduped and I want my rows back
What you see: The rows are gone after you click the red button, and Undo does not bring them back.
Why it happens: Dedupe deletes the rows. It is not on the Undo list, it leaves no line in History (Change log or Restore data), and nothing appears in the Trash.
What to do: Add the rows again from your source. Next time, check the column and the list before you click the red button.
Good to know
Dedupe deletes rows. Check which column you use first.
Rows removed by Dedupe are not on the Undo list, and they leave nothing in History or the Trash. Treat the delete as permanent.
Only the first occurrence of each value is kept.
Blank values are not listed, and the blank rows stay.
See What can I get back? Undo, History, Archive and Trash for what you can and cannot bring back.