Batches

You can create batches in Relativity by splitting a static set of documents into multiple document sets based on criteria set by a system admin. With the appropriate permissions, users can then check out these batches and assign themselves documents.

Batching can bring structure to managing a large-scale document review with multiple reviewers. You can generate batches automatically or manually. Automatic document assignment reduces the chances of accidental redundancy in the review process.

You may want to consider using Review Center as a streamlined alternative. For more information, see Review Center.

Batch fields in Relativity

The fields in Relativity that store batch-related values are:

• Batch Set—the batch set to which the batch belongs.
• Batch—the name of the batch.
• Batch Status—the status of the batch. A batch may be:
  • Blank—the batch is not checked out.
  • In progress—the batch is checked out.
  • Completed—the batch is checked in as completed.
• Batch Unit—the optional grouping condition on the batch.
• Assigned To—the reviewer to whom the batch is checked out.
• Reviewed—specifies whether the batch is reviewed.
• Batch Size—the number of records in the batch.

Creating and editing batch sets

Documents are pre-converted for viewing upon the checkout of a batch set based on the NumberOfDocumentsToAutoPreconvert instance setting. This setting is set to 1000 by default and can be adjusted to suit your needs.

If the value is set to 0, pre-conversion will be disabled.

To create or edit a batch set, perform the following steps:

1. Navigate to the Batch Sets tab.
2. Click New Batch Set to create a new batch set, or click Edit next to the batch you want to edit.
   If you select a value in the Family Field, then the batch set will include family documents from that related group that are also included in the data source.
   The Batch Set Information form appears with required fields in orange.
   Complete the fields to create or edit the batch set. See Fields.
3. Click Save.

Batch Sets for Reviewers

Once you have created the batches, navigate to the Batch Sets tab using the quick search, CTRL + /, function. Here you will see a reviewer's view of the newly created batches.

Fields

The following fields display on the New Batch Set information form:

• Name—the name of the batch set.
  If you are performing a multi-stage review with multiple batch sets per phase, be sure to include the phase name in the batch set name. We also do not recommend using the space bar before or after the name. This makes it easier to search for information within a specific phase.
• Maximum Batch Size—the maximum number of documents in one batch. If you also select a Family Field option, the Maximum Batch Size value may be exceeded. To learn more, see Exceeding the maximum batch size below.
  Your batch may contain fewer documents than this value if there are no more documents to batch or no more documents with the current Batch Unit value.
• Batch Prefix—sets the prefix for the batch numbering.
• Batch Data Source—the saved search containing documents that you want to batch. If the saved search references a dtSearch or an Analytics index, that index must be active. The sort order of the data source can influence how documents are ordered in batches and how the documents are grouped for batching.
• Batch Unit Field—an optional grouping mechanism, enabling you to group similar documents in the same batch. Documents with different values in a batch unit field are not included in the same batch unless the Family Field would override this.
  A common batch unit field is custodian. However, you can use any single choice, multi-choice, single object or non-relational and non-system fixed length text field.
  When using multiple choice fields for the Batch Unit Field, the choice values must be the same in order for them to be grouped in a batch. If the combination of choices is unique enough for the choice values, Relativity batches the documents into separate batch groups.
• Family Field—you can select your workspace group identifier. Documents in a family group are not split across batches, even if their batch unit field differs.
  If you select an option for this field, the Maximum Batch Size may be exceeded. To learn more, see Exceeding the maximum batch size below.
• Reviewed Field—an optional field that you can select from any yes/no, single-choice, or multi-choice field from the drop-down menu. Based on your selection, a tally is kept of how many documents in the batch have been reviewed. The batch set monitors the field you specify as the Reviewed field. It also increments a count for each document in the batch that has been coded with a value for that field.
• Auto Batch—enables the system to automatically generate batches.
  • Disabled—the default, which enables standard manual batch processing. 
  • Enabled—automatically generates batches according to the settings below.
• Minimum Batch Size—only available if Auto Batch is enabled. This is the smallest possible number of documents in a batch.
• Auto Create Rate—only available if Auto Batch is enabled. This represents how often, in minutes, the system attempts to create batches.
  Newly created batches appear on the Review Batches tab. Any users with access to the Review Batches tab will be able to see any batches that they have permissions for.
  With auto batching enabled, the Relativity service account runs the saved search specified as your batch data source. The Relativity Service account is a system admin account.

Batch Set console

Once you save your batch set, the Batch Set console displays the following options:

• Create Batches—creates the batches based on the entered settings. Created batches display on the bottom half of the batch set details page.
• Purge Batches—deletes any batches associated with the entire batch set.
• View Batch Summary Report—displays the Batch Summary report. This field is only enabled after you have created batches.
• Refresh Page—refreshes the page.

 Auto Batching Status

This section displays on a batch after it has been created to help you track the status of auto batching. The following options display:

• Status—displays the current status of the batch: Pending, Processing, Error or Finished
• Documents to be Batched—displays the remaining documents that do not meet the minimum batch size. You must manually create a batch for these remaining documents.
• Last Successful Run—the date and time stamp of the last successful batch run.
• Last Error Reported—the date and time stamp of the last error that occurred.

Deleting an individual batch

To delete an individual batch using the Delete mass operation, do the following:

1. Navigate to the Review Batches tab.
   
2. Select the individual batch you would like to delete from the list.
3. Select Delete from the mass operations drop-down menu in the bottom-left.
   
   A pop-up window displays.
4. Click Delete in the pop-up.
   The selected batch is removed.

Deleting a batch set

Use the following procedure to delete a batch set.

1. Navigate to the Batch Set tab.
2. Select the checkbox next to the batch you want to delete.
3. Select Delete in the mass operations drop-down menu, and click Go.
   A confirmation message appears.
4. Click OK.

For more information, see Mass delete.

Searching for documents not in existing batch sets

In some cases, you may need to search for all documents not included in a batch set. To properly search for documents not included in a batch set, use the following search criteria:

• Field: Batch
• Operator: not these conditions
• Value:
  • Field: Batch::Batch Set
  • Operator: any of these
  • Value: select all existing batch sets

Documents included in the batch sets you select for the Batch::Batch Set field value are excluded from your search results.

Batching process overview

The batching process works in a loop which contains the following steps:

1. Identify the next set of documents to be placed into a batch, based on the batch set's configuration, the documents returned by the saved search, and the list of documents that have already been placed in batches in the batch set.
   The process that this step follows can vary depending on whether the batch set is configured with a Batch Unit Field or Family Field. The potential variations are described in the sections below.
2. Create a new batch and assign those documents to the batch.
3. Determine if there are documents returned by the saved search that have yet to be placed into a batch associated with the batch set. If there are unassigned documents, this loop starts over again from the first step.

No Batch Unit Field or Family Field configured

• Relativity searches for the next set of document returned by the saved search that have not been placed into a batch yet. These documents are used to populate the next batch.
  • The number of documents in the next batch will be between 1 and the Maximum Batch Size field value.

Batch Unit Field configured; No Family Field configured

1. Relativity searches for the next single document returned by the saved search that has not been placed into a batch yet.
2. Relativity searches for the next set of documents returned by the saved search that have not been placed into a batch yet and have a Batch Unit Field value that matches the Batch Unit Field value of the document found in step 1. These documents are used to populate the next batch.
   • The number of documents in the next batch will be between 1 and the Maximum Batch Size field value.

Batch Unit Field and Family Field configured

1. Relativity searches for the next single document returned by the saved search that has not been placed into a batch yet.
2. Relativity searches for the next set of documents returned by the saved search that have not been placed into a batch yet and have a Batch Unit Field value that matches the Batch Unit Field value of the document found in step 1.
   • The number of documents will be between 1 and the Maximum Batch Size field value and these documents will include between 1 and Maximum Batch Size field value of unique Family Field values.
   • Each document with an empty Family Field value is treated as if it is in a unique family by itself.
3. Relativity searches for all documents returned by the saved search that have not been placed into a batch yet and have a Family Field value that matches the Family Field value of any of the documents found in step 2. These documents are used to populate the next batch.
   • This process does not take Batch Unit Field into consideration and at minimum, all documents that were identified in the previous step are included in the batch.
   • Each document with an empty Family Field value is treated as if it is in a unique family by itself. Even if one or more of the documents returned in the second step had an empty Family Field value, documents with empty Family Field values that have Batch Unit Field values that do not match the Batch Unit Field value of the document identified in step one will not be included in these results.

No Batch Unit Field configured; Family Field configured

1. Relativity searches for the next set of documents returned by the saved search that have not been placed into a batch yet.
   • The number of documents will be between 1 and the Maximum Batch Size field value and these documents will include between 1 and the Maximum Batch Size field value of unique Family Field values.
   • Each document with an empty Family Field values is treated as if it is in a unique family by itself.
2. Relativity searches for all documents returned by the saved search that have not been placed into a batch yet and have a Family Field value that matches the Family Field value of any of the documents found in step 1. These documents are used to populate the next batch.
   • Each document with an empty Family Field value is treated as if it is in a unique family by itself.

Exceeding the maximum batch size

When a batch set has a Family Field value, it is possible for the number of documents in individual batches to exceed the Maximum Batch Size value. This is due to the fact that the batching process first identifies the next documents to be placed in a new batch based on the Maximum Batch Size and Batch Unit Field values before proceeding to include all documents returned by the saved search that have a Family Field value that matches the Family Field value of any document initially identified as a candidate for the new batch and have not yet been placed into any batches belonging to the relevant batch set. For more information about the batching process, see the Batching process overview section above.

Based on the document data set in the workspace, the selected saved search, and the batch set field values, it may not be possible to completely prevent this from happening. For example, a family that includes more documents than the Maximum Batch Size value, will always result in at least one batch within the batch set that exceeds the Maximum Batch Size limit.

Depending on the scenario, it may be possible to reduce the number of occurrences where the Maximum Batch Size is exceeded by adjusting the sort option on the saved search which is used as the Batch Set Data Source. Families that tend to be clustered together within the results of a saved search or within the results of a saved search as grouped by Batch Unit Field values will more often naturally fall into the same batch, based on the other batch set field values. Sort options that cause family documents to be split and spread out across the results of a saved search will increase the likelihood that these documents will end up being batched together in a way that they exceed the configured maximum batch size. Choosing saved search sort options that prioritize clustering families of documents together, can sometimes help reduce the occurrences of this issue.