> ## Documentation Index
> Fetch the complete documentation index at: https://docs.worksong.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Export a board

> Export a board, or one view of it, to Excel, CSV or JSON, and bring a JSON export back as a new board.

You can export a board, or just the view you are looking at, to a file. Work builds the file in the background, so a large board never holds you up. You get a notification with a link when it is ready.

## Export

Open the board's options and choose **Export**. Pick a format:

| Format | Good for |
| - | - |
| Excel (`xlsx`) | A spreadsheet. One sheet for the items, and an optional sheet for comments. |
| CSV | UTF-8 with a byte-order mark, so Excel reads it. One file is one table, so it cannot include comments. |
| JSON | A full copy of the board, which you can import back as a new board. |

You can export the whole board or only the view you are looking at (its filter and its visible columns). You can also include comments (Excel and JSON) and archived items.

You need Read on the board. An export is private to the person who asked: not even the Owner can open someone else's. **Column permissions apply.** A column you cannot see is not in the file, in any format.

## What is in the file

* Excel and CSV have a `Group` column first. People appear by name, Connect values by the linked item's title, Link as `label (url)`, Checkbox as Yes or No, and a Timeline as `start to end`.
* A cell that starts with `=`, `+`, `-` or `@` is written as text and never run as a formula. In CSV it gets a leading apostrophe.
* A Button column and a Waits for column are empty in the file. A Time tracking column is the hours of its stopped sessions, for example `1.5`.
* Subitems are not exported yet, and neither are their comments.

## Limits

* A file holds up to 50,000 items, 2,000,000 cells, 100,000 comments and 64 MiB. Over a limit the export fails and tells you what to filter.
* You can have 3 exports being prepared at once. Work keeps your newest 5.
* A file can be downloaded for 7 days.
* Your access is checked when the file is built. If you lost access in between, the export fails and no file is made.

## Import a JSON export

`POST /v1/imports/board-json` creates a **new** board from a JSON export. You need Manage. Use `?mode=preview` first: it writes nothing and shows the board name, columns, row count, skipped columns and dropped cells. Then use `?mode=commit`.

* **Comes back:** the name, and the Text, Long text, Status, Person, Date, Number, Link, Multi-select, Checkbox, Tags, Email, Phone, Rating, Timeline, Location, World clock and Hour columns with their settings, and every item's cells.
* **Does not come back:** Formula, Lookup, Progress, Connect, File, Doc, AI, Button, Waits for, Time tracking, Vote, Created at and Updated at columns. Groups, comments and item ids are not imported either. Every item lands in the first group. A cell whose value is not valid here, such as a person who is not in this workspace, is dropped and counted.
* At most 5,000 items and 100 columns per import. Export a filtered view to import a bigger board in pieces.
* A refused file answers `422` with a `code`, such as `invalid_json`, `not_a_board_file` or `too_many_items`.

## API

| Call | Notes |
| - | - |
| `POST /v1/boards/{id}/exports` | Starts an export and answers `202`. Body: `format`, and optionally `query`, `tz`, `columnIds`, `includeUpdates`, `includeArchived`. |
| `GET /v1/boards/{id}/exports` | Your newest exports, at most 20. |
| `GET /v1/boards/{id}/exports/{exportId}` | Status: `queued`, `running`, `done` or `failed`. |
| `GET /v1/boards/{id}/exports/{exportId}/file` | The file. It needs the same bearer token as every call, so fetch it with the header. |

<sub>Source: `worksong-ai/worksong:services/work/docs/board-export.md@956ab3b`. Docs version: Worksong 2.0 with Work 1.2.</sub>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.