---
title: "Split text into columns in Strata | Oplema"
description: "Split comma-separated or other delimited text into columns with Split to columns in Strata. Choose the separator and whether to keep the original column."
url: "https://oplema.com/docs/strata/parse/split-to-columns"
updated: "2026-11-04"
---

# Split to columns

Separate text such as comma-separated names or pipe-separated codes into columns. Choose the separator and keep each part on the same row, with the option to keep the original text.

**Learn this in context:** [Split packed cells into columns or rows](https://oplema.com/learn/many-sources/split-packed-cells) (Combine and reconcile data files)

## Split names into columns

Manual Input has four rows with id and name columns. Split to columns starts with name selected and a comma separator. Ada,Lovelace becomes Ada in name\_1 and Lovelace in name\_2.

Result
| id | name | name\_1 | name\_2 |
| --- | --- | --- | --- |
| 1 | Ada,Lovelace | Ada | Lovelace |
| 2 | Lin,Chen | Lin | Chen |
| 3 | Sam | Sam |  |
| 4 |  |  |  |

1. Clear Keep original column. The `name` column disappears from the output, leaving `id`, `name_1` and `name_2`.
2. Change Splitter to `Pipe (|)`. None of the names contain a pipe, so the entire name goes into `name_1` and `name_2` disappears.
3. Choose `Custom` in Splitter and enter a comma in Pattern. The names split into two columns again.
4. Use Reset example, then select the Select block and select its output port. It renames `name_1` to `first_name` and `name_2` to `last_name`, keeping the original values.

## When to use

Use Split to columns when one cell holds a fixed set of parts, such as last name and first name separated by a comma. If the number of items varies, use Split to rows.

## Read the output

With the default settings, the output contains `id`, `name`, `name_1` and `name_2`. Ada's row becomes `Ada` and `Lovelace`; Lin's becomes `Lin` and `Chen`. The original names stay alongside the new columns.

The row containing `Sam` has no comma. Its `name_1` is `Sam` and `name_2` is an empty string. The final row has a NULL name, so both split columns are NULL. An empty string is a text value; NULL means a value is missing.

All four rows stay. Clearing **Keep original column** only removes `name` from this block's output; it doesn't change the input table.

The example ends with [Select](https://oplema.com/docs/strata/prep/select), which gives the numbered columns meaningful names: `first_name` and `last_name`. With the original settings, its output contains `id`, `name`, `first_name` and `last_name`.

## How many columns are created?

The value with the most parts sets the number of output columns. They are named after the source column, followed by `_1`, `_2` and so on, and are added after the kept input columns.

Rows with fewer parts get empty strings in the remaining positions. NULL source values produce NULL parts. If no value contains the separator, the block creates just one split column containing the original values.

To put each part on a separate row instead, use [Split to rows](https://oplema.com/docs/strata/parse/split-to-rows).

## Inputs and outputs

### Inputs

1 port

**In Required**

Accepts: Tables, Geospatial data.

Accepts one connection on this port.

### Outputs

1 port

**Out**

## Configure the block

Choose the text column and the literal separator between its parts.

**Column**

Choose the text column to split. Only text columns are available; the first is selected automatically.

**Splitter**

Choose Comma (,), Pipe (|), Tab, Semicolon (;), Space or Custom. The separator is matched exactly and is removed from the resulting parts. Default: Comma (,).

**Pattern**

Shown only for Custom. Enter a nonempty literal separator, such as ::. It can contain multiple characters and isn't a regular expression.

**Keep original column**

Keep the unsplit text alongside the new columns, or clear to remove it from the output. Other input columns remain. Default: true.

## Split to columns FAQs

**Does splitting add or remove rows?**

No. Every input row stays on one output row. The split parts become additional columns.

**Are spaces trimmed from the parts?**

No. Spaces stay unless they're part of the separator. Splitting Ada, Lovelace on a comma leaves a leading space before Lovelace.

**What happens with consecutive separators?**

They produce empty parts. For example, Ada,,Lovelace creates Ada, an empty string and Lovelace in three numbered columns.

**What if an output column name already exists?**

The block reports an error rather than overwriting it. Rename the existing column before splitting.

### Prepare your data. Build it in Strata.

Strata opens in your browser, with no account.

[Open Strata](https://oplema.com/editor?path=%2F%3Fopen%3Dstrata) [About Strata](https://oplema.com/strata)
