> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://support.atomicwork.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://support.atomicwork.com/_mcp/server.

# Importing records via CSV

> Bulk-load and update records in any custom object from a CSV file using a guided three-step import.

When you already have your data in a spreadsheet — a vendor list, a contract register, a client roster — you don't need to recreate it record by record. You can import it into any custom object from a CSV file and update those records later by re-uploading the same file.

Use a CSV import when you want to:

* **Load hundreds of records in one upload** instead of creating them one at a time.
* **Update existing records** by re-uploading the same file with changed values.
* **See who imported what and when** from the import history.

---

## Import records

1. Go to **Directory > \[Object name]**.
2. Click **Import**, then select **Import via CSV**.
3. Work through the three steps of the guided import: **Details**, **Map attributes**, and **Validate and review**.

### New and existing records

Each new row in your file creates a new record. If your file also contains records that are already in Atomicwork, tick **Overwrite data for existing objects** to update those matching records from the same file.

### Map attributes

Atomicwork automatically maps your CSV columns to the object's attributes. Review the mapping and adjust any column that landed on the wrong attribute before you continue.

### Validate and review

Atomicwork validates your file before it writes anything. You'll see a summary such as "15 rows are ready to import, no errors found."

Required fields are enforced during validation, so every row needs a value for each required attribute. Optional fields can be left blank.

---

## Track your imports

The import history lists every file imported into the object. For each import, you can see:

* **Status** — In progress, Completed, Completed with errors, or Failed
* **Records imported** — how many records were imported from the file
* **Errors** — how many rows had errors

We'd recommend checking the import history after any import that shows **Completed with errors**, so you can correct your file and re-upload it.

---

## Related

* [Creating Object Types](/documentation/assets-objects/custom-objects/build-the-model/creating-object-types)
* [Fields & Sections](/documentation/assets-objects/custom-objects/build-the-model/fields-sections)