aiR Custom Analysis (Advanced Access)

Advance Access (AA) is an opportunity to evaluate and work with Relativity features prior to the General Availability (GA) release. Relativity customers typically participate in AA programs on a feature-by-feature basis. The functionality described in this topic may not be available in all Relativity environments. This topic may not represent the functionality, appearance, or behavior of the GA release version of this feature.

Overview

We’re excited to invite you to participate in the Advanced Access (AA) program for Custom Analysis; a new RelativityOne capability that introduces unprecedented flexibility to document review and extensibility to additional legal data intelligence (LDI) use cases.

By participating in the AA program, you agree to test the Custom Analysis framework and provide Relativity with feedback so we can continue to build and improve upon this feature.

Related topic:

• Best practices guide for Custom Analysis field descriptions (Advanced Access)

The Custom Analysis Framework

Custom Analysis introduces a flexible framework for designing and executing custom AI‑powered insights within Relativity. Instead of relying solely on predefined analysis types, users can create up to five Custom Analysis fields per project, each acting as an independent analysis for questions, extractions, classifications, or evaluative instructions applied to every eligible document. Custom Analysis supports both Text and Vision analysis, with each Custom Analysis field running independently across eligible documents.

Reviewers will receive per‑document outputs, dependent on their Custom Analysis prompt; delivering on the creativity and bespoke needs of our clients.

AA Program Plan

• AA Program Duration: The program will run from late March 2026 through late June 2026. During this time, participants will have early access to aiR Custom Analysis and the opportunity to provide feedback that will help shape the final release. Participants will have continuous access to Custom Analysis throughout the AA window.
• Deployment: The feature will be deployed to participating workspaces at the start of the AA program. There will be a 1,000 document limit per job and total 25,000 document limit per customer, and participants will not be billed.
• Data Requirements: Custom Analysis supports both Text and Vision analysis, with each Custom Analysis field running independently across eligible documents. Supported image types include native image files (JPEG, PNG, GIF).

Feedback

Your feedback during the AA program is critical to the continued development and refinement of aiR Custom Analysis. You may receive requests for feedback with your Relativity AA program contacts to share your experience and observations. During the program, we encourage you to focus your feedback on some of the following areas: Custom Analysis accuracy and relevance, impact on review efficiency, user experience and usability, and overall value and usefulness.

Your input will directly inform our roadmap and help ensure the feature delivers meaningful value upon general availability.

Help and Support

If you have further questions about how this feature works or if you have issues with the feature, please contact Bryce Lew (bryce.lew@relativity.com) or Salma Kamni (salma.kamni@relativity.com).

Key Features Available in the Custom Analysis Advanced Access Program

The Custom Analysis Advanced Access Program allows users to define and manage up to five custom fields per project. Each field acts as an independent instruction, enabling customizable results across all your LDI use cases.

Custom Fields

These instructions form the core logic for what the model returns for each document. Every Custom Analysis field consists of two user‑defined components:

• Field Name—A required, unique label (up to 50 characters) that appears as a column header in the Results table and document viewer.
• Description / Prompt—A required instruction (up to 500 characters) that tells aiR how to analyze the documents and how to present the results. Examples include summarizing information, extracting items, identifying attributes, or applying classifications. These instructions form the core logic for what the model returns for each document. For guidance on writing effective prompts, see Best practices guide for Custom Analysis field descriptions (Advanced Access).

Behavior and Output

Each field runs independently, with no cross‑field influence during AA. For every document, each Custom Analysis produces one text‑based output, which may include:

• Summaries
• Yes/No responses
• Classification labels
• Short explanations
• Identification

Managing Custom Analysis Fields

Users can adjust Custom Analysis fields at any point before running or re‑running analyses. Supported actions include:

• Editing prompt descriptions
• Deleting fields
• Viewing the Custom fields in the Viewer panel next to your documents.

Feature documentation

Installation

Ensure you have the aiR for Review application installed in the workspace(s) where you will be testing.

Configuration

Custom Analysis introduces a flexible but structured configuration experience that allows users to define Custom Analysis fields, set up project parameters, run analyses, and manage versions. The setup process is designed to support rapid experimentation while ensuring accuracy, defensibility, and repeatability.

The configuration workflow consists of four primary steps:

1. Setting Up a Project

2. Creating and Managing Custom Analysis Fields

3. Running an Analysis Job

4. Viewing the Results

Setting Up a Project

Begin by setting up the aiR for Review project as described below.

1. Navigate to the aiR for Review Project tab.
2. Click New aiR for Review Project.
3. Fill out the following fields on the Setup Project modal:
   
   • Project Name—enter a name for the project.
   • Description—enter a useful short description of the project.
   • Data source—select the saved search that holds your document sample set. Refer to Choosing the data source for more information.
   • Analysis Type—select Custom.
   • Project Prompt Criteria—select one of the following:
     • Start blank—select this if you plan to write new prompt criteria from scratch.
     • Copy existing—select this to use a previously created set of prompt criteria in this workspace and copy it into this project for modifying as needed.
   • Prompt Criteria Name—either keep the default name that automatically populated or click the pencil icon to rename it. This name must be unique.
   • Custom Analysis Model—this field displays when Custom is selected for the Analysis Type. Select the analysis model desired:
     • Text—run analysis on the Extracted Text of documents.
     • Vision—run analysis on documents where their native file type is either JPEG, GIF, or PNG.
   • Project Use Case—choose the option that best describes the purpose of the project. If none of the options describe the project, choose Other and type your own description in the field. It will only be used for this project. Keep this description generic and do not include any confidential or personal information. The Project Use Case field is used for reporting and management purposes and does not affect how the project runs.
4. Click Create Project.

Creating and Managing Custom Analysis Fields

Within the Custom Analysis panel, you can define up to five Custom Fields that determine how aiR will analyze each document. Each field is created and edited directly within the prompt criteria pane.

Custom Field Limits

• Custom Fields—each Custom Analysis project can include up to five Custom Fields.
• Field Title (Name)—the field name is required. It must be unique within the project and can be up to 50 characters. The title will act as the label for displaying results in Analysis Results tables, Custom Projects card in the Viewers, and the published output.
• Description / Prompt—the description or prompt is required. It can be up to 500 characters. It defines the exact instruction given to the model for its analysis. See Best practices guide for Custom Analysis field descriptions (Advanced Access) for more information.

Custom Analysis Field Management

On the Custom Analysis panel you can create new Custom Analysis fields, edit existing field names and prompts, and delete fields that are no longer needed.

1. Enter a unique Title (Name) for the first Custom Field (up to 50 characters). It will act as the label for displaying results in Analysis Results tables, Custom Projects card in the Viewers, and the published output.
   
2. Enter clear and descriptive instructions in the prompt text box (up to 500 characters) that will guide aiR in analyzing documents.
3. Click + Field to add more Custom fields and repeat the above steps for each one. A maximum of five Custom Fields can be entered.
   
4. To delete a custom field from the list, click the vertical ellipses next to the desired field and select Delete.
5. Click Save when finished adding Custom fields.

Running an Analysis Job

Once custom field setup is complete, you can run the analysis.

1. Click Analyze [X] documents.
2. Review the confirmation summary modal showing the total number of documents to be analyzed.
3. Click Start Analysis.

Each custom field runs independently across eligible documents. Documents that cannot be analyzed will display with errors.

Viewing the Results

Once the analysis completes, Custom Analysis results display as columns in the analysis results table and in a dedicated Custom Projects card within the viewer. If users modify custom fields, change the data source, or create a new versioned set, only the updated fields will re-run on the next analysis.

Frequently Asked Questions

Q: Can I publish my results to the workspace?

A: During the AA period, publishing results is not currently supported. We anticipate this being available by the end of April.

Q: How much does it cost to run a Custom Analysis job?

A: During the AA period, you will not be charged, nor incur any additional aiR Unit usage.

Q: How can I provide feedback, or ask questions?

A: You can contact Bryce Lew (bryce.lew@relativity.com) or Salma Kamni (salma.kamni@relativity.com).