To generate an Agent Auto Complete evaluation, the interaction must meet the following minimum requirements:

• The transcript must contain at least 300 characters

If either of these conditions is not met, Genesys Cloud does not generate an evaluation for the interaction.

These requirements ensure that enough interaction data is available to generate an accurate evaluation.

Programs support the automated generation of AI-driven, auto-complete evaluations.

1. Open the relevant program.
2. Click the Agent Scoring tab.
3. Click Add Rule.
4. Configure the scoring rule:
   • Interaction volume
     • All interactions – Generates an evaluation for every interaction that matches the program.
     • Percentage of interactions – Generates evaluations for a specified percentage of matched interactions. Valid values range 1–100%.
   • Agent auto-complete evaluation form – Select a published agent auto-evaluation form.
5. Save the rule.
6. Enable the rule and publish the program.

You can configure up to 20 scoring rules per program. Scoring rules are optional and can be enabled or disabled at any time.

For more information, see Work with a program.

You can generate an evaluation programmatically by calling the Evaluations API.

Endpoint

Required fields

• conversationId
• evaluationFormId
• evaluationFormType = "Automated"
• agentId (optional)

Example request body

1. From the Form list, select an Agent Auto-Complete evaluation form.
2. Set the Evaluator type to Auto Submit with Virtual Supervisor.
3. Click Create.