> ## Documentation Index > Fetch the complete documentation index at: https://parabola.io/docs/llms.txt > Use this file to discover all available pages before exploring further. # Merge duplicate > The **Merge duplicates** step allows you to group rows in one or more columns and merge their values. ## Input/output The data we'll input into this step has an identifier type to deduplicate and its corresponding columns of data. Input table with multiple rows sharing the same identifier value The step provides us with output of deduplicated data (based on a unique identifier value) with the merged values separated by a **Delimiter** in their corresponding columns. Output table deduplicated by identifier with merged values separated by a delimiter ## Default settings By default, Parabola will map your first column to the **unique value** selector, and your second column to your **Merge** selector. Default Merge duplicates settings with the first column mapped as the unique identifier The **Delimiter** will be automatically set to a comma (,). Delimiter field set to a comma by default Lastly, the option to **Ignore blanks** will be auto-selected on. ## Custom settings First, decide whether you want to include or exclude columns, by selecting **these columns** or **all columns except**. Setting to include or exclude columns from the merge using these columns or all columns except Next, click the plus sign to select the columns you would like to merge. Adding columns to merge by clicking the plus icon and selecting from the column list Then, use the **Per unique value in** selector to set the column which will act as your unique identifier. Per unique value in selector specifying which column acts as the unique identifier You can choose whatever **Delimiter** you want. By default, it is a comma (,) but you can choose whatever works best with your data, such as a space ( ), a dash (-), or more. Custom delimiter field accepting comma, space, dash, or other characters Selecting the option to **Only merge unique values** keeps all values unique. For example, differently formatted phone numbers will not roll up into one merged number. Only merge unique values option enabled to keep distinct values from being collapsed **Ignore blanks** is selected by default and treats blank cells as if they are not data points. Unchecking this option includes blank cells into your merged values, separated by your **Delimiter**. Ignore blanks checkbox controlling whether blank cells are included in merged values ## Related steps * [Remove duplicate](/product/transform/remove-duplicate-rows) — drop duplicate rows entirely instead of merging their values. * [Sum by group](/product/transform/sum-by-group) — total numeric columns by a grouping key. * [Combine columns](/product/transform/combine-columns) — concatenate columns within a single row. * [Split column](/product/transform/split-column) — reverse a merge by splitting delimited values back apart. * [Pivot columns](/product/transform/pivot-columns) — pivot duplicate row values into columns instead.