> ## Documentation Index
> Fetch the complete documentation index at: https://koreai.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Interaction Evaluation and Conversation Analytics Report

[Back to Reports List](https://koreai.mintlify.site/ai-for-service/analytics/quality-ai/reports/qm-reports-list)

The Interaction Evaluation & Conversation Analytics Report provides interaction-level insights for quality evaluation and business analysis across customer conversations. It exports interaction metadata together with evaluation outcomes, sentiment analysis, detected topics, audit information, AI-generated justifications (when available), conversation intelligence insights, business metadata, evaluation metric details, and any selected optional or custom fields.

The report includes evaluation outcomes, sentiment analysis, detected topics (including topic hierarchy where available), metric-level evaluation details, manual audit results, AI-generated justifications, business metadata, and conversation intelligence insights.

Use filters such as channel, queue, flow, language, agents, agent groups, contact direction, and duration status to generate targeted reports. By default, exports include only processed conversations, with optional inclusion of pending and validation-failed conversations. Optional and custom fields are added as CSV columns based on the selected evaluation forms, report configuration, and available data.

## Generate the Report

> [!NOTE]
>
> The report includes only interactions evaluated using the selected evaluation forms.

New report configuration:

1. **Create the New Report**.

   Follow the standard report creation process described in see [Scheduled Reports](https://koreai.mintlify.site/ai-for-service/analytics/quality-ai/reports/scheduled-reports#create-a-report).

2. Select **Interaction Evaluations and Conversation Analytics Report** from the **Report Type** list.

   ![Report Selection](https://mintcdn.com/koreai/N4xuxqRvqrsh2OGV/ai-for-service/analytics/quality-ai/reports/images/interaction-report.png?fit=max\&auto=format\&n=N4xuxqRvqrsh2OGV\&q=85\&s=2181a9b3ce337851b489feea07b8a8bb)

3. Select an **Evaluation Form** (at least one form is required). Only interactions evaluated using the selected forms are included in the report. The available evaluation data and optional fields can depend on the selected evaluation forms.

4. Select a **Date** or **Date Range** and **Timezone**.

5. Complete the **Configuration** — **Filters and Options** section that defines the interaction data included in the report.

6. Configure **Filters**. The report supports **Standard Filters** and **Custom Filters**.

   * **Standard Filters**: Select **+ Add Filter** to add one or more standard filter rows to configure the required criteria. This enables you to use predefined system attributes for consistently segmenting and analyzing interactions across reports.

     Available filters can include: **Agents**, **Agent Groups**, **Language**, **Channel**, **Contact Direction**, **Flow** (when applicable to the selected evaluation form), **Duration Status**.

     The exact filter options available can vary based on the selected evaluation form and the data available for the selected date.

     Configure the selected filter values. When using a flow-based evaluation form, the Flow filter displays only the Experience Flows associated with the selected evaluation form.

   * **Custom Filters**: Select **Custom Filters** using organization-specific fields. This is provided by CRM integrations, APIs, AI Agent metadata, or other organization-specific configuration.

     Set filter values, and segment interaction data by attributes such as `AccountID`, `AccountStatus`, `AccountStatusID`, `ActionID`, `ActionName`, `Content ID`, `Phone number`, `agentEmail`, and other configured custom fields. These filter field names and available values are dynamic and depend on the organization's configured data sources.

   * Select **Optional Fields** to include additional metrics or attributes in the report export that provide extended interaction performance and customer context. Only the optional fields selected in this section are added to the report output. Optional Fields are organized into two tabs:

     * **Standard Optional Fields**: These are predefined, system-defined fields ready to use for consistent reporting. Depending on the selected evaluation forms and available data, these can include: **Custom Conversation ID**, **Agent Groups**, **End Time**, **Duration**, **Empathy Score**, **Crutch Word Score**, **Intents**, **Keywords**, **Topics**, **Agent Emotions**, **Is Churn Risk Observed**, **Evaluation Form**, **Flow Name**, **Conversation Source**, **Agent AI Widget Status**, **Metric Violation Events**, **AI Justification**, **Audit Status**, **Audit Date**, **Audit Response by Matrix**, **Disposition**, **Evaluation Form Scoring Type**, **Not Applicable Scoring Treatment**, **Context**, **Comments** ], and other available conversation intelligence metrics.

     * **Custom Optional Fields**: These are organization-defined fields made available through CRM integrations, APIs, AI Agent metadata. Examples can include **Content ID**, **Phone Number**, **agentEmail**, **boolean check**, **Customer Name**, **emailcheck**, **Queue**, **queueId**, **JSON Check**.

       The actual fields available for selection depend on the organization's configuration, the selected evaluation forms, the available data, and the selected date range.

7. Enter one or more email addresses in the **Send Email to**. At least one email address is required.

8. Select **Next**.

9. Complete the **Schedule** section to generate the report using the current configuration.

> [!NOTE]
>
> AI Justifications are available only for **Gen AI by Question** metrics, provide AI-generated explanations for each evaluation score. They appear in **Interaction Evaluations and Conversation Analytics** reports, and are available through APIs.

For information about report configuration, see [Scheduled Reports](https://koreai.mintlify.site/ai-for-service/analytics/quality-ai/reports/scheduled-reports#configure-schedule).

> [!NOTE]
>
> The report returns all interactions matching the selected filters. Use the **Duration Status** filter to limit results by evaluation eligibility. Interactions ingested before duration threshold support may contain a `null` value for `duration_status`.

## Processing Status Options in Reports

The following options determine whether pending or validation-failed conversations are included in exported reports.

| Option | Description |
| - | - |
| **Include pending conversations** (disabled by default) | Exports conversation metadata for pending conversations. Evaluation scores and Conversation Intelligence fields remain empty until processing completes. |
| **Include validation-failed conversations** (disabled by default) | Exports conversation metadata together with the validation failure reason. |

> [!NOTE]
>
> Exported reports include a **Processing Status** column with one of the mentioned values: **Processed**, **Pending**, or **Validation Failed**. For validation failure, the report also includes the corresponding validation failure reason.

***

# Report Behavior

The report behaves as follows:

* AI Justifications are available only for **Gen AI by Question** metrics.
* Custom fields depend on the selected date range and available data.
* Only selected optional and custom fields are included in the CSV export.
* By default, all interactions matching the selected filters are exported.
* Use **Duration Status** to filter interactions by: **Evaluated**, **Below Threshold**, **Duration Unavailable**.
* For interactions ingested before duration threshold support, the `duration_status` field returns `null`.
* Contact Direction displays **Inbound**, **Outbound**, or **All Directions**, when available.
* For flow-based evaluation forms, the report includes Experience Flow conversations and exports flow-specific evaluation details.
* The **Not Applicable Scoring Treatment** field reflects the treatment (Exclude from Score or Award Full Weightage) applied to that interaction's evaluation at the time it was scored.
* When **Process Conversations When AgentAI Widget Is Off** is turned on for the Agent AI source, the report includes Agent AI interactions processed with the widget turned off. Select **Agent AI Widget Status** under **Standard Optional Fields** to identify them.

***

## Report Fields

### Generated CSV Fields

The generated CSV exports one row for each interaction and includes interaction metadata, evaluation outcomes, sentiment analysis, detected topics, audit information, AI-generated justifications (when available), conversation intelligence insights, business metadata, evaluation metric details, and any selected optional or custom fields.

| Field | Description |
| - | - |
| **Account ID** | Account identifier associated with the interaction. |
| **Contact ID** | Unique identifier for the customer contact. |
| **Conversation ID** | Unique identifier assigned to the interaction. |
| **Entry Date** | Date and time when the interaction started. |
| **Channel** | Communication channel used for the interaction, such as Voice or Chat. |
| **Contact Direction** | Indicates whether the interaction is Inbound, Outbound, or All Directions. |
| **Queues** | Queues associated with the interaction. |
| **Flow Name** | Experience Flow associated with the interaction. Available only for flow-based evaluation forms. |
| **Conversation Source** | Optional. Platform that ingested the interaction, such as CCAI, Agent AI, or Automation AI. |
| **Agent AI Widget Status** | Optional. Indicates whether the AgentAI widget was Enabled or Disabled for the interaction. Available only for Agent AI interactions. |
| **Processing Status** | Indicates whether the interaction is Processed, Pending, or Validation Failed. |
| **Duration Status** | Indicates whether the interaction is Evaluated, Below Threshold, or Duration Unavailable. |
| **Evaluation Scores** | AutoQA and manual evaluation scores for the interaction. |
| **Last Queue** | Final queue that handled the interaction. |
| **Language** | Configured or detected interaction language. |
| **Agents** | Agents associated with the interaction. |
| **Scored Agent** | Agent whose performance is evaluated. |
| **Start Time** | Date and time when the interaction started. |
| **End Time** | Date and time when the interaction ended, when selected as an optional field. |
| **Sentiment** | Conversation sentiment identified during analysis. |
| **Topics** | Detected conversation topics (L1, L2, and L3 where available). |
| **Escalation** | Indicates whether the interaction was escalated. |
| **Kore Evaluation Score** | Overall Quality AI evaluation score. |
| **Is Passed** | Indicates whether the interaction passed evaluation. |
| **Is Fatal Interaction** | Indicates whether the interaction contains a fatal metric violation. |
| **Audit Score** | Manual audit evaluation score. |
| **Not Applicable Scoring Treatment** | Indicates whether Not Applicable metrics are excluded from the score or awarded full weightage (Exclude from Score or Award Full Weightage) for this interaction's evaluation. |
| **Contact Direction** | Inbound, Outbound, or All Directions. |
| **Duration Status** | Indicates whether the interaction met duration requirements for evaluation. |
| **Evaluation Form** | Evaluation form used for the interaction. |
| **Call Transfer Update** | Result of the Call Transfer metric. |
| **Manual Evaluation UK1 - Manual Evaluation Metric** | Result of the manually evaluated UK1 metric. |
| **Playbook UK** | Result of the Playbook UK metric. |
| **Speech Crosstalk Update** | Detects overlapping speech during voice interactions. |
| **Speech Dead Air** | Detects excessive silence during voice interactions. |
| **Speech Rate Violation** | Evaluates whether the speaking rate meets configured standards. |
| **AI Justifications** | AI-generated explanations for Gen AI by Question metric scores, where available. |
| **Conversation Intelligence** | Conversation intelligence insights included in the report. |
| **Optional Fields** | Additional columns selected during report configuration. |
| **Custom Fields** | Organization-specific fields included in the export. |
| `<Metric Name>` | Evaluation result for the configured quality metric. |
| `<Metric Name> - Is Fatal Metric` | Indicates whether the metric is configured as fatal (`1 = Yes`, `0 = No`). |
| `<Metric Name> - Adherence Weight` | Weight assigned to the metric during score calculation. |
| `<Metric Name> - Violation` | Indicates whether the interaction violated the metric (`1 = Violation`, `0 = No Violation`, `NA = Not Applicable`). |
| `<Metric Name> - AI Justification` | AI-generated explanation supporting the metric result, when available. |
| `<Metric Name> - Trigger Justification` | Explanation of the detected trigger or condition that produced the metric result. |

> [!NOTE]
>
> The exported CSV dynamically includes columns based on the selected evaluation forms, applied filters, optional fields, custom fields, and report configuration. For flow-based evaluation forms, the export also includes the **Flow Name** column where applicable.

***

### Metric-Level Evaluation Fields

When metric-level evaluation details are included in the export, the report contains one row for each evaluated metric associated with an interaction.

| Field | Description |
| - | - |
| **Metric Name** | Name of the evaluation metric. |
| **Is Fatal Metric** | Indicates whether the metric is configured as fatal (`1 = Yes`, `0 = No`). |
| **Total Interactions** | Total number of interactions evaluated for the metric. |
| **Total Applicable Interactions** | Number of interactions in which the metric was applicable. |
| **Total Adhered Interactions** | Number of applicable interactions meeting the evaluation criteria. |
| **Metric Adherence %** | Percentage of applicable interactions that adhered to the metric. |
| **Comments** | Comments entered by the auditor for the corresponding manual audit metric result, when available. |

For manual evaluation metrics, auditor comments are associated with the individual metric result. They aren’t equivalent to AI Justification or Trigger Justification.

Metric-level fields are generated dynamically based on the selected evaluation form. Each configured metric includes its evaluation result together with related attributes such as fatal metric status, adherence weight, violation status, and applicable auditor comments.

Supported metrics may include: Standard quality metrics, AI-based metrics, Deterministic metrics, Manual evaluation metrics, Voice quality metrics, Organization-specific custom metrics.

AI Justification and Trigger Justification fields are included only for supported metrics.

> [!NOTE]
>
> Metric evaluation fields are repeated for each evaluated metric associated with an interaction. A single interaction may appear in multiple rows of the exported CSV — one row for each evaluated metric.

***

### View and Download the Report

For information about viewing, downloading, managing reports, and adding, editing, or removing filters, see [Manage Reports](https://koreai.mintlify.site/ai-for-service/analytics/quality-ai/reports/scheduled-reports#manage-reports).

Reports are exported in [CSV format](https://raw.githubusercontent.com/Koredotcom/docs-v2/refs/heads/main/ai-for-service/assets/ie-report.csv).

#### Field Representation in the Generated CSV

The generated CSV exports one row for each interaction and includes interaction details, evaluation outcomes, conversation intelligence insights, and any selected optional or custom fields.

**Typical exported information includes**: Interaction identifiers and timestamps, channel and contact direction, queue and Flow Name (for flow-based evaluation forms), processing status and duration status, evaluation outcomes (including evaluation scores, pass or fail status, Not Applicable Scoring Treatment, and fatal interaction indicators), metric-level evaluation details, sentiment analysis and detected topics, AI-generated justifications (when applicable), evaluation metric details (if configured), conversation intelligence insights, selected optional fields, and custom business fields.

> [!NOTE]
>
> Additional evaluation metric columns are included when supported by the selected evaluation form.

The exported CSV includes interaction-level details followed by metric-level evaluation details for each applicable evaluation criterion.

***


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.