Agent auto evaluation form

AI-generated summary

Genesys Cloud's Agent Auto-Evaluation Forms feature automatically submits evaluations once all questions are answered through AI Scoring, Evaluation Assistance, or predefined default responses, enabling scalable quality monitoring programs with reduced evaluator workload. The feature requires Genesys Cloud CX 1 WEM Add-on II or higher licensing, along with specific permissions for evaluation form management and AI scoring configuration. AI Experience Tokens are required within Genesys Cloud CX environments. Configuration requires administrators to select a scoring method for each question and establish fallback default answers. AI Scoring uses Genesys Cloud AI models to determine answers based on conversation transcripts and metadata, while Evaluation Assistance applies predefined rules and conditions such as keywords or topics. Default answers are applied when AI Scoring confidence scores fall below 70% or when Evaluation Assistance conditions are not met. All questions must be assigned either AI Scoring or Evaluation Assistance mechanisms, with every question group requiring a configured default answer. Evaluation Assistance questions must include at least one condition. Forms without assigned scoring types cannot be published. When evaluations are created, the system automatically submits them and designates the evaluator as Virtual Supervisor. Evaluations can be created through API integration using POST endpoints for specific interactions or recurring evaluations across programs. Agreement scores should be reviewed to identify questions with low reliability, as these may indicate unclear wording or automation challenges requiring refinement. The platform supports multiple user roles including agents, administrators, developers, partners, supervisors, and support personnel. Additional resources and community support options are available for users requiring assistance with evaluation form creation, publishing, and generation workflows.

PrerequisitesClick to expand

In Genesys Cloud CX, Agent Auto-Evaluation Forms are automatically submitted once all questions have been answered—whether through AI Scoring, Evaluation Assistance, or predefined default responses. This automation accelerates feedback cycles, reduces evaluator workload, and enables organizations to scale their quality monitoring programs with minimal manual effort.

Key benefit: Automatically submit agent evaluations—without human review—when all questions are resolved by AI or default responses.

How it works

When building an Agent Auto-Evaluation Form, administrators must:

Select a scoring method for each question—AI Scoring or Evaluation Assistance.
Configure a fallback default answer for every question.

The default answer is used when the selected scoring method cannot determine an answer:

- For AI Scoring questions, the default answer is selected when the interaction is flagged as low accuracy or when the model confidence score is below 70%. For more information, see Automated scoring options in evaluation forms.
- For Evaluation Assistance questions, the default answer is selected when the configured conditions are not met and no answer can be selected.

When an evaluation is created using an Agent Auto-Complete evaluation form, every question is answered automatically. The system then submits the evaluation, listing the evaluator as Virtual Supervisor.

Scoring types

Each form question must use one of the following mechanisms:

AI Scoring – Uses AI models in Genesys Cloud AI to determine answers based on conversation transcripts and metadata.
Evaluation Assistance – Applies predefined rules and conditions (such as keywords or topics) to suggest answers.

Important: Each question must use either AI Scoring or Evaluation Assistance. If a question does not have one of these scoring types, a configuration error will prevent the form from being published.

Validation rules

To enable auto-submission, the form must meet the following criteria:

All questions must use AI Scoring or Evaluation Assistance as the scoring method.
Every question group must have a configured default answer.
If Evaluation Assistance is selected, each question must include at least one condition.

Note:

To create evaluations for a specific interaction using an Agent Auto-Complete evaluation form, use the POST /api/v2/quality/conversations/{conversationId}/evaluations API or create a new ad-hoc evaluation.
To create recurring evaluations with an Agent Auto-Complete evaluation form using rules, use the POST /api/v2/quality/programs/{programId}/agentscoringrules API.
For more information about how to best design, test and fine tune an auto complete evaluation, see AI Scoring best practices.

For Agent Auto-Complete evaluation forms, review the agreement score for each question in the latest published version of the form. Questions with low agreement scores may indicate unclear wording, incomplete scoring guidance, or answer options that are difficult to automate. Improving these questions can increase the reliability of automated scoring and reduce the likelihood that questions are excluded or flagged because of low historical accuracy. Questions with low agreement scores may also be more likely to be affected by the historical accuracy filter and, as a result, may not be answered automatically. For more information about the historical accuracy filter, see Automated scoring options in evaluation forms.

For more information about automated scoring question options, see Automated scoring options in evaluation forms.
For more information about how to generate evaluations that use Agent Auto-Complete evaluation form type, see How do I generate an Agent Auto-Complete Evaluation?

[NEXT] Was this article helpful?

Get user feedback about articles.