> ## Documentation Index
> Fetch the complete documentation index at: https://docs.databar.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Deduplication
> Remove duplicate rows from your tables, manually or automatically.
Databar gives you three ways to remove duplicates from your tables: per-column deduplication, a one-off full-table cleanup, and automatic deduplication that prevents duplicates as new data arrives.
## Deduplicate by a single column
You can remove duplicates based on one specific column directly from the column header.
1. Click on the **column header** you want to deduplicate by.
2. Select **Remove duplicates** from the context menu.
Databar removes all rows that have duplicate values in that column, keeping only the first occurrence.
This operation is **irreversible**. It permanently deletes duplicate rows from your table.
## Remove all duplicates at once
You can also run a one-off deduplication check across all columns in your table.
1. Click the **Settings** button in the table toolbar.
2. Under the **De-duplication** section, click **Remove duplicates now**.
This checks every column in the table and removes any rows that are full duplicates across all columns.
## Auto de-duplication
Auto de-duplication runs continuously in the background and prevents duplicates from being added to your table in the first place. You can turn it on and off at any time.
To set it up:
1. Click the **Settings** button in the table toolbar.
2. Click **Auto de-duplication**.
3. Toggle **Automatic de-duplication** on.
4. Select the columns you want to check for duplicates. You can pick multiple columns.
5. Click **Save**.
### How auto de-duplication works
When you save your auto de-duplication settings, Databar immediately runs a "remove all duplicates" operation on the selected columns to clean up any existing duplicates in the table. From that point on, it checks incoming data on a rolling basis before it enters the table. If a new row matches an existing row on any of the selected columns, it is blocked from being added.
The order of operations matters: the table checks for duplicates first, cancels any duplicate rows, and only then runs enrichments on the remaining new rows. This means you never waste credits enriching data that would have been removed as a duplicate.
### When to use auto de-duplication
* Tables that receive continuous data from webhooks, imports, or scheduled runs
* Lead lists where you want to avoid contacting the same person twice
* Any table connected to an automated pipeline where duplicates could accumulate over time
## Comparing the three options
| Method | Scope | When it runs | Removes existing duplicates |
| --------------------- | ---------------- | ------------ | --------------------------------- |
| Column de-duplication | Single column | On demand | Yes |
| Remove all duplicates | All columns | On demand | Yes |
| Auto de-duplication | Selected columns | Continuously | Yes (on save) + prevents new ones |
## Related
Learn how tables work in Databar
Bring external data into your tables