> For the complete documentation index, see [llms.txt](https://docs.docbits.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.docbits.com/administration-and-setup/settings/global-settings/document-types/table-columns/purpose-and-use.md).

# Purpose and Use

A table column is one field of the line-item table. Everything that DocBits does with a table (extraction, validation, PO matching, export) works on the columns configured here.

## Where a column shows up

| Place                   | What the column does there                                                                                                                                          |
| ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Validation screen**   | One column in the line-item table. The *Title* is the header, the *Column Type* decides the editor (amount, date, text, yes/no). Hidden columns are not shown.      |
| **Table training**      | When you train a supplier's table, you map each detected table column to one of these configured columns. Only configured columns can be mapped.                    |
| **AI table extraction** | The AI fills the configured columns. A column marked *Use AI* is filled by the AI even for suppliers with trained rules.                                            |
| **Validation rules**    | Line-item checks such as *quantity × unit price = line total* run on the default columns `QUANTITY`, `UNIT_PRICE`, `TOTAL_AMOUNT`, `CHARGES`, `DISCOUNT`.           |
| **PO matching**         | Needs the default columns item number, unit price, quantity and total amount. Without them the document shows *Line Item Table is missing Mandatory column for PO*. |
| **Export**              | Every non-hidden column is part of the line-item data sent to the ERP. The export mapping references the *Column name*.                                             |
| **Scripts**             | Scripts read and write columns by *Column name*, for example `row["TOTAL_AMOUNT"]`.                                                                                 |

## Scope

* Table columns are configured **per table**, and a table belongs to a **document type**. Invoice columns do not affect delivery notes.
* The configuration is **per organization**. Sub-organizations inherit it.
* Which columns are *filled* for a given supplier is decided by that supplier's training or by the AI; the column configuration only says which columns exist.

## Typical reasons to change the configuration

* A customer-specific value must be captured per line (cost centre, project number, internal article number) → add a column.
* A default column is never used and clutters the validation screen → hide it.
* A column must always be filled before export → mark it *Required*.
* A value comes from the ERP lookup and must not be edited by users → mark it *Read Only*.
* The AI captures a column better than the trained rules (for example free-text descriptions) → mark it *Use AI*.

## Related pages

* [Adding a new column](/administration-and-setup/settings/global-settings/document-types/table-columns/adding-a-new-column.md)
* [Editing and deleting columns](/administration-and-setup/settings/global-settings/document-types/table-columns/editing-and-deleting-columns.md)
* [Table Extraction Troubleshoot](/overview-and-basics/faq/document-processing/table-extraction-troubleshoot.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.docbits.com/administration-and-setup/settings/global-settings/document-types/table-columns/purpose-and-use.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.