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

# Importing Data

> Upload CSV files with automatic schema detection

<Frame>
  <img src="https://mintcdn.com/prismeai/9aPAXX2QoCsMDyNB/images/ai-collection-create.png?fit=max&auto=format&n=9aPAXX2QoCsMDyNB&q=85&s=5c1978bce49cc84a27531ae381e93b1b" alt="Create Collection dialog with CSV import option" width="1574" height="840" data-path="images/ai-collection-create.png" />
</Frame>

The fastest way to populate a collection is to upload a CSV file. Collections parses your file, detects column names and types automatically, and shows you a preview before anything is created.

## Uploading a CSV

<Steps>
  <Step title="Start a new collection">
    Click **New Collection**, enter a name and optional description, then select **Upload CSV**.
  </Step>

  <Step title="Add your file">
    Drag your `.csv` file into the drop zone, or click it to open the file picker. Files up to **50 MB** are accepted.
  </Step>

  <Step title="Review the detected schema">
    Once parsed, each column appears as a badge with its detected type, for example `company: text`, `deal_value: number`, `close_date: date`.
  </Step>

  <Step title="Check the data preview">
    A paginated preview table shows sample rows exactly as they will be stored. Page through it to spot formatting issues before committing.
  </Step>

  <Step title="Create">
    Click **Create**. The collection is created with the detected schema and all rows are imported. Use the **Remove** (trash) icon at any point before creating to discard the file and start over.
  </Step>
</Steps>

## How Type Detection Works

Collections scans a sample of your rows and assigns each column the most specific type that fits all its values:

| Detected type | Recognized values             | Example                          |
| ------------- | ----------------------------- | -------------------------------- |
| `number`      | Integers and decimals         | `1247`, `3.14`                   |
| `date`        | ISO dates and datetimes       | `2026-02-25`, `2026-02-25 14:30` |
| `boolean`     | `true` / `false` (any casing) | `TRUE`, `false`                  |
| `url`         | `http://` or `https://` links | `https://example.com`            |
| `text`        | Everything else               | `Acme Corp`                      |

A few rules to know:

* **Mixed columns fall back to text.** If a column contains `12`, `15`, and `N/A`, it becomes `text`; clean the stray values first if you need numeric aggregations.
* **Empty cells are allowed.** Columns with missing values are marked nullable; empty cells are shown as a dash.
* **Numbers are strict.** Only plain decimal notation is parsed (no scientific notation, no `0x` prefixes).

<Tip>
  For dates, export in ISO format (`YYYY-MM-DD`) from your source tool. Localized formats like `25/02/2026` are not recognized as dates and will import as text.
</Tip>

## File Requirements

| Requirement       | Value                                             |
| ----------------- | ------------------------------------------------- |
| Format            | `.csv` (comma-separated, standard RFC 4180)       |
| Maximum file size | 50 MB                                             |
| Header row        | Required; the first row provides the column names |
| Encoding          | UTF-8 recommended                                 |

Excel files (`.xlsx`) are not supported directly; export your sheet as CSV first.

## Import Errors

| Message                    | Cause                                                       | Fix                                    |
| -------------------------- | ----------------------------------------------------------- | -------------------------------------- |
| *Not a CSV file*           | The file extension or content is not CSV                    | Export/save as `.csv`                  |
| *File exceeds 50 MB limit* | The file is too large                                       | Split it or remove unused columns      |
| *No data found in CSV*     | The file is empty or only has a header                      | Check the export from your source tool |
| *Failed to parse CSV file* | Malformed rows (unbalanced quotes, inconsistent separators) | Re-export with standard settings       |

## Adding Data Later

Importing at creation is not the only way to fill a collection:

* **Agents can write records** during conversations, using the collection's insert and update tools; see [Using Collections with Agents](/products/ai-collection/agents)
* **The API accepts batch inserts** of up to 1,000 records per call; see [API Reference](/products/ai-collection/api)

Schemas are flexible: records added later can introduce new fields, and the collection's columns adapt without a migration.

<Note>
  **Available soon**: the capability store will add **connectors** (Salesforce, HubSpot, and more) to sync data into collections directly from your business tools, and **hooks** to pre-process records as they arrive.
</Note>

## Next Steps

<CardGroup cols="2">
  <Card title="Working with Data" icon="table" href="/products/ai-collection/data-management">
    Browse and manage your imported records
  </Card>

  <Card title="Using Collections with Agents" icon="robot" href="/products/ai-collection/agents">
    Let agents read and write your data
  </Card>
</CardGroup>