> ## Documentation Index
> Fetch the complete documentation index at: https://koreai.mintlify.app/llms.txt
> Use this file to discover all available pages before exploring further.
# By Dialog Task Metric
The By Dialog Task metric evaluates how consistently agents follow predefined scripts, workflows, and behavioral steps during customer interactions. It enforces best practices, supports compliance, and drives high-quality service outcomes.
## Why Use This Metric
Configuring By Dialog Task metrics lets you:
* Audit performance at the entire playbook level or drill down into specific steps.
* Set adherence thresholds to flag unacceptable variances.
* Improve agent performance and customer experience continuously.
* Identify training or coaching needs based on non-compliance behaviors.
* Ensure consistency across agents and communication channels.
* Support regulatory and procedural compliance in regulated industries.
* Configure time-bound evaluation windows for timely feedback.
## Use Cases
| Use Case | Description |
| ----------------------------------- | --------------------------------------------------------------------------- |
| **Compliance Monitoring** | Evaluate mandatory process steps in regulated industries (finance, telecom) |
| **Onboarding and Escalation Flows** | Help new agents follow onboarding scripts or escalation workflows |
| **Quality Assurance Reviews** | Audit specific conversation parts (greetings, disclaimers, closings) |
| **Training Needs Analysis** | Identify agents or teams that frequently deviate from prescribed scripts |
| **Performance Benchmarking** | Compare adherence scores across agents, shifts, or teams |
| **Customer Experience Improvement** | Detect and correct deviations that affect customer satisfaction |
***
## Prerequisites
Before configuring By Dialog Task metrics, confirm:
1. **Playbooks or dialog tasks exist** — define dialog tasks or playbooks before creating this metric.
2. **Configuration access** — you must have permissions to access **Quality AI** > **Configure** > **Evaluation Metrics**.
3. **Evaluation forms defined** — if you plan to use metrics in forms, configure forms beforehand.
4. **No active dependencies for deletion** — metrics linked to active evaluation forms or attributes cannot be deleted. Reassign or remove dependencies first.
5. **Clear evaluation criteria** — choose the correct Measurement Type (Entire Conversation or Time Bound). For time-bound analysis, define boundaries in seconds (Voice) or message counts (Chat).
***
## Configure By Dialog Task Metric
1. Navigate to **Quality AI** > **Configure** > **Evaluation Forms** > **Evaluation Metrics**.
2. Select **+ New Evaluation Metric**.
3. From the **Evaluation Metrics Measurement Type** dropdown, select **By Dialog Task**.
4. Enter a descriptive **Name** for future reference.
5. Select a **Dialog Agent** from the dropdown.
6. Select a **Count Type**:
| Option | Description |
| ----------------------- | -------------------------------------------------------------- |
| **Entire Conversation** | Checks adherence throughout the entire interaction |
| **Time Bound** | Checks adherence within a specific time range or message count |
For **Time Bound**, configure:
* **Parameter** — select First Part or Last Part of the conversation.
* **Voice** — enter the number of seconds from the start or end.
* **Chat** — enter the number of messages from the start or end.
7. Select **Create**.
***
## Edit or Delete By Dialog Task Metric
1. Select the By Dialog Task metric in the evaluation metrics list.
2. Select the option:
* **Edit** to modify the metric.
* **Delete** to remove the metric.
3. Select **Update** to save changes.
### Deletion Requirements
Before deleting a metric:
* Remove it from all associated evaluation forms.
* Reassign any linked attributes to a different metric.
* The system allows deletion only after all dependencies are resolved.
***