---
title: "Collection"
description: "Before you begin collecting files from a custodian's data source, you must create a collection. You must name the collection, as well as associate it with a specific matter, custodian, and one or more data sources. You can later use this information to generate various reports that characterize the data retrieved from the custodian's machine."
feature: "Collect"
audience: ["end user", "developer", "system analyst", "system administrator", "system engineer"]
version: "Relativity aiR"
type: ["User Guide", "Tutorial"]
url: "https://help.relativity.com/RelativityaiR/Content/Collect/Collection/Collection.htm"
lastModified: "2026-10-29"
---

# Collection

Before you begin collecting, you must create a collection job and associate it with a specific matter, custodians, and one or more data sources. Add the custodians, data sources, and other information using the Collect wizard. Once completed, start the collection using the Collect console. Finally, download a results report that details the items collected and a summary report of the entire collection job. For more information, see[Reports](https://help.relativity.com/RelativityaiR/Content/Collect/Collect_reports.md).

## Considerations

Before creating a collection for any data source, consider the factors that can affect collection performance.

- Collection time can increase when you use:
  - Large date ranges.
  - A high number of custodians.
  - Multiple collections running at the same time.
- Combining large scopes with high concurrency can slow down collections. This behavior is expected across all data sources.
- When you migrate a Collect workspace, data source credentials do not move with the workspace. You must re-enter credentials before running new Collect jobs. For more information, see [Migrate](https://help.relativity.com/RelativityaiR/Content/Collect/Collection/Migrate.md).

## Best practices for time‑sensitive collections

To improve performance and reduce collection time:

- Use smaller date ranges whenever possible.
- Limit the number of custodians included in each collection.
- Reduce concurrency, or run collections sequentially when appropriate.
- Create multiple collection jobs when necessary.

Following these practices can help collections complete more quickly and predictably.

## Creating a collection

Before you begin creating a collection, make sure to create a matter. For more information, see [Matters](https://help.relativity.com/RelativityaiR/Content/Collect/Matters.md).

Use the following procedure to create a collection:

1. On the Collect tab, click the **Collections** sub-tab. Collect displays a list of the collections currently added to this application.
2. Click **New Collection**.
3. Complete the steps in the Collect wizard. See [Using the Collect wizard](#Using)
4. On the Collection Details page, click **Run Collection** in the console. See [Collect console](https://help.relativity.com/RelativityaiR/Content/Collect/Collection/Collection_details.md#CollectConsole).

For information on running concurrent Microsoft 365 collection jobs, see [Accessing Microsoft 365 tenants](https://help.relativity.com/RelativityaiR/Content/Collect/Data_sources/Microsoft_365_-_OneDrive.md#AccessingM65Tenants).

## Using the Collect wizard

The Collect wizard takes you through each step to create a collection. After completing the collection setup, run a collection from the Collection console.

**Collection wizard security permissions**

- **Custodian**—View

When a step is complete, click **Next** or the hyperlink under the next step shown. Click the **Previous** button to move to the previous step. Relativity auto-saves your information when moving between steps. If any required information is wrong or missing, an error message displays and you cannot move to the next step.

### Collection Details

Complete the Collection Details step by entering information in the following fields:

- **Name**—the name of the collection. Enter a name using alphanumeric characters only. You cannot use special characters, such as periods, commas, and em dashes. Special characters will cause an error.
- **Collection Matter**—the name of the matter associated with this collection. Click **Edit** to select an existing matter or click **Add** to define a new one. See [Creating a matter](https://help.relativity.com/RelativityaiR/Content/Collect/Matters.md#Creating).
- **Job Number**—lists a number assigned to the job for reporting purposes.
- **Processing Source Location**—the file repository for collected data to be stored for future processing of documents or for storing collected data. All data sources that produce RSMF have a limit of 2 GB. For more information, see [Processing](https://help.relativity.com/RelativityaiR/Content/Relativity/Processing/Processing_in_Relativity.md).
- **ZIP Collected Files**—toggle on to compress all collected data into ZIP64 formatted containers. Relativity can compress Microsoft 365, Slack, and X1 data into zip folders.
  - Selecting *Yes* on the Collect Files in ZIP field adds your collected data into containers and puts it into a password protected compressed folder. The compressed folders separated by the custodian target collected. Each custodian target collect has its own folder. These folders will split when reaching a set size. The compressed folders are then stored in the processing staging area by default.
  - You can secure these compressed folders with a password. You can enter a password that you, or another user, needs to enter to open the compressed folder. Relativity stores these passwords in the password bank. You can retrieve them there at a later time.
  - To export your collected data, use the Staging Explorer. For information, see [Staging Explorer](https://help.relativity.com/RelativityaiR/Content/Relativity/RelativityOne_Staging_Explorer/RelativityOne_Staging_Explorer.md).
  - Google compresses the data on their end and they exclude Google data from the ZIP64 formatted containers.
  - iManage does not support adding data to ZIP64 formatted containers.
- **Zip Password**—enter a password that is required by anyone attempting to decompress the ZIP64 container files.
  - If you have Processing installed in the workspace, Collect automatically populates the Processing Password Bank with the password so it is available at the time the collection is processed.
  - Click the Show Password box to display the actual password in the field to ensure the characters are correct.
  - Zip is supported by Microsoft 365, Slack, and X1. Google is exported as pre-zipped.
- **Enable Auto-Processing**—toggle on to enable auto-processing. If enabled, select the workspace, profile, and document prefix. Relativity processes all data in a completed collection, or completed with errors collection, after a collection finishes.
  - **Workspace**—select a workspace within your instance to use when creating a processing job.
  - **Processing Profile**—select a processing profile available in the drop-down menu. The available profiles are from the selected workspace.
  - **Document prefix** —select the document prefix option of *Use Entity Document Number Prefix* or *Use Processing Profile Document Number Prefix* to apply to each file in the processing set once it's published to a workspace.
- **Description**—enter a description of the collection used for reporting purposes.
- **Receive Progress Notifications**—toggle on to send or receive collection job status emails. The statuses include:
  - **Completed**—includes completed or completed with errors job status.
  - **Failed**—includes job status and reason for failure.
- **Notification Address**—enter the email address of person that wants to receive collection job statuses.
- **Data Source Type**—select one or more data sources to use in the collection. For more information, see [Data source types](https://help.relativity.com/RelativityaiR/Content/Collect/Data_sources/Data_sources.md#Data).

### Data source

Configure the data sources chosen in the Collection Details step. Each data source has different criteria to enter. See the list of data source types under [Data sources](https://help.relativity.com/RelativityaiR/Content/Collect/Data_sources/Data_sources.md) for information on each one.

You can select multiple data sources in the first step if you want to configure all or multiple sources in the step. Switch between each source to configure its criteria by using any of these methods:

- Click the name of the data source in the navigation menu.
- Click **Next** and **Previous** to move you through the data sources.
- Select individual data sources by clicking on the checkbox and then using the right arrows to select them.

#### Data source criteria

Add criteria to collect specific data. To configure the data sources, complete the following fields:

- **Select and unselected tabs**—choose the data sources to collect from by moving unselected data sources to the selected list.
- **Field**—choose the field to filter on within the data source.
- **Operator**—choose an operator such as equals, contains, greater than, or less than.
- **Value**—enter a value to find in the selected field.

After selecting field options, you must click **Add Criteria**. You can add multiple criteria to search data sources.

Note the following about criteria:

- Each criteria is then separated by an AND operator.
- Leave the data source criteria empty to collect all data from the sources.

### Custodians

Complete the Custodians step by assigning custodians to the project. Follow the steps below to assign a custodian.

1. From the Unselected custodians table, use the column filters to locate custodians.
2. Select a checkbox next to a custodian.

   > **Notes:** If you select multiple Slack custodians in the same collection, the channels they share will only collect once.

   - Collect supports up to 30 custodians, or entities, assigned in a collection job.
   - If you need more than 30 custodians, you need to create another collection job.
   - This limit only applies to the number of custodians.
   - There is no limit to the number of targets to be collected.
   - For example, you can select 30 custodians and three data sources for a total of 90 targets for the collection job.
3. Click the right arrow ![Right Add Arrow icon](https://help.relativity.com/RelativityaiR/Content/Resources/Images/Features/Relativity_Legal_Hold/10.0/ICN_RightAddArrow.png) icon to add select custodians. Click double right arrow ![Right Add All arrow icon](https://help.relativity.com/RelativityaiR/Content/Resources/Images/Features/Relativity_Legal_Hold/10.0/ICN_RightAddAllArrow.png) icon to add all custodians.
4. Click **Next**.

> **Note:** There is a limit of 10,000 listed custodians with targets in the custodian picker.

### Non-custodial

> **Note:** This section only applies when SharePoint or Smarsh (Advanced Access) is selected as the data source.

Select non-custodial data sources to complete this step. Non-custodial data means you select the specific locations or archives you want to collect from. Selecting custodians is not required.

To select non-custodial data:

1. Click one or more data sources in the **Select Sources** column.
2. With one of the data sources highlighted, select the check boxes next to the sites you want to collect.
   1. **SharePoint**—you can collect from parent-level sites only. All sub-sites under a parent site are automatically collected.
   2. **Smarsh**—Smarsh organizes data into cases. Select the cases you want to collect from. Collect only supports collection jobs of one case for Smarsh.
3. (Optional) Toggle on the **Show selected only** option to display only the locations selected for collection. The selected locations display in a concise list for review before you proceed to the Summary step.
4. Click **Next**.

### Collection Summary

Complete the creation of the collection by reviewing all steps, custodians, data sources, and targets, before finalizing.

If Google Workspace, Microsoft 365, Slack, and X1 custodian targets were not created before you started the project, click **Generate Targets**.

Click **Generate Targets** to check if targets exist for the custodians you have selected for collection. If the targets do not exist, Collect automatically creates them based on the email address contained in the Entity record for each custodian.

#### Targets

In the Targets section, a number appears next to each custodian's name. The number listed is the number of custodian targets found in the associated data source.

| Number | Meaning |
| --- | --- |
| 0 | Collect did not find any custodian targets with that email address in that data source. |
| 1 | Collect found a single custodian target associated with the email address within the data source. |
| 2+ | Collect found multiple custodian targets with that email address within the data source. |

The number is also color-coded to indicate how the target was generated:

- **No highlight**—Collect already found and generated the custodian target.
- **Green**—Collect autogenerated the custodian target.
- **Red**—Collect could not autogenerate the custodian target.
  - You can still manually generate it.
  - For more information, see [Creating a custodian target](https://help.relativity.com/RelativityaiR/Content/Collect/Custodian_targets.md#Creating).

Click **View Collection Details** to finish setup. The Collection Details page appears. From the Collection Details page, you can preview and run the collection from the [Collection Summary](#Collecti).

#### Non-custodian targets

> **Note:** This section only applies when SharePoint is selected as the data source.

The *Non-Custodian Targets* section displays the target, target identifier, and the status.

- **Target**—the name you gave when creating the data source.
- **Target Identifier**—the URL of the SharePoint site.
- **Status**—a message telling you if the target is valid or invalid. If invalid, navigate to the data source details page and click the **Refresh Sites** button.
