---
title: Handling Enhanced Data Stores
description: How to create Enhanced Data Stores, upload and manage files, reindex documents, search within a store, and delete stores or individual files.
---

# Handling Enhanced Data Stores

The [News → Data Sources → Enhanced Data Stores]({APP_HOST}/documents/pipelines) page lists all Enhanced Data Stores in your organization. From here you can create new stores, browse their contents, manage files, and delete stores.

See [Enhanced Data Stores](enhanced-data-stores) for a conceptual overview of what they are and how they work.

## Creating an Enhanced Data Store

1. Go to [News → Data Sources → Enhanced Data Stores]({APP_HOST}/documents/pipelines).
2. Click **Create**.
3. Work through the setup wizard.

### Step 1 — Choose a Content Source

Select where documents will come from:

- **User Files** — You upload files manually. The store is empty to start with; you add files after creation.
- **Latest News from a Data Source** — The store fetches the most recent articles from an existing online data source. You select the source and an optional cutoff date.
- **Search Results from a Data Source** — The store runs a keyword query against an existing data source and indexes the results. You provide the search term and select the source.

For the last two options, proceed to Step 2 to select the data source.

### Step 2 — Select the Data Source *(online sources only)*

Pick an existing online data source from your organization. Only sources that support the required capability for the chosen content type (Latest or Search) are shown.

### Step 3 — Choose a Transform

Select how documents should be processed before indexing. The **Default** transform is the right choice for most scenarios — it auto-detects the format and extracts text. For specialized content types (audio, video, scanned PDFs), choose the appropriate transform.

See [Enhanced Data Stores → Transforms](enhanced-data-stores#processing-pipeline) for descriptions of all available transforms.

### Step 4 — Choose a Splitter

Select how documents are divided into chunks. **Paragraph stacking** is the default and works well for most documents. Choose **Whole document** for very short files, or **Hierarchical** for long structured documents with clear headings.

See [Enhanced Data Stores → Splitters](enhanced-data-stores#processing-pipeline) for descriptions of all splitter options.

### Step 5 — Choose an Index Technology and Embedding Model

Select **Flash** (recommended) as the index technology. Then pick an embedding model from the list. The embedding model determines how text is converted into semantic vectors; choose a model that supports the primary language of your content.

### Step 6 — Name the Store

Enter a name and click **Create**. The store is created and appears in the list. If you chose **User Files**, you can now open the store and start uploading.

## Uploading Files

1. Open an Enhanced Data Store of type **User Files** from the [News → Data Sources → Enhanced Data Stores]({APP_HOST}/documents/pipelines) list.
2. Click **Upload** (or **Files** to go to the file management view).
3. Click **Upload** on the files page, or drag and drop files onto the page.
4. Select one or more files from your computer.

Files are uploaded immediately. Each file then enters a processing queue and progresses through these statuses:

| Status | Meaning |
|---|---|
| **Waiting** | The file has been uploaded and is waiting to be processed |
| **Processing** | The transform and splitter are being applied |
| **Ready** | Processing is complete; the file is indexed and searchable |
| **Error** | Processing failed — hover over the status to see the error message |

The page refreshes automatically every 10 seconds while files are being processed, so you can monitor progress in real time.

## Searching Within an Enhanced Data Store

Once files have the **Ready** status, you can search within the store:

- From the **Enhanced Data Stores** list page, click **Search** next to the store.
- From the files view of a User Files store, click **Search** in the toolbar.

Both links take you to [News → Search]({APP_HOST}/documents/search) with the Enhanced Data Store pre-selected as the source.

## Reindexing Files

Reindexing re-processes a file with a (possibly different) transform and splitter and rebuilds its index entry. Use reindexing when:

- You want to apply a different transform (for example, switching to audio transcription after uploading a podcast).
- A file failed processing and you want to retry.
- You have updated the splitter settings and want to rebuild all chunks.

### Reindex a single file

1. Open the files view of the Enhanced Data Store.
2. Click the **Reindex** button (sync icon) next to the file.
3. In the dialog, optionally change the transform and/or splitter.
4. Click **Reindex**.

### Reindex multiple files

1. Select the files you want to reindex using the checkboxes.
2. Click **Reindex (N)** in the selection toolbar.
3. In the dialog, optionally change the transform and/or splitter.
4. Click **Reindex**.

After reindexing starts, the affected files revert to **Waiting** status and are processed again.

## Downloading Files

To download a file from a User Files store:

1. Open the files view.
2. Click the **Download** button (down-arrow icon) next to the file.

The file is downloaded to your computer using its original file name.

## Deleting Files

### Delete a single file

1. Open the files view.
2. Click the **Delete** button (bin icon) next to the file.
3. Confirm the deletion.

### Delete multiple files

1. Select the files you want to remove using the checkboxes.
2. Use **Select all** in the toolbar to select every file in the current view if needed.
3. Click **Delete (N)** in the selection toolbar.
4. Confirm the deletion.

Deleted files are removed from both the file store and the search index. This action is permanent.

## Renaming an Enhanced Data Store

1. Go to [News → Data Sources → Enhanced Data Stores]({APP_HOST}/documents/pipelines).
2. Click **Rename** next to the store.
3. Enter the new name and click **Save**.

## Deleting an Enhanced Data Store

Deleting a store removes it along with all its files and index data.

1. Open the files view of the store you want to delete.
2. Click **Delete** in the toolbar.
3. Confirm the deletion.

You can also delete a store from inside its files view. Deletion navigates you back to the Enhanced Data Stores list.

> **Warning:** Deleting an Enhanced Data Store is permanent. All uploaded files and indexed content are removed and cannot be recovered.
