Quality Analysis: Categories
- Product Knowledge
On this page |
FROM VERSION 3.36.0
Overview
Categories are sets of filters designed to help analyze voice recordings within Post Call Analytics.
By default, all created categories are automatically applied to transcripts generated through Call Analytics.
Categories must be created before submitting a transcript request. They cannot be applied retroactively.
Each category consists of 1 to 20 rules, which define the specific conditions that trigger category detection during analysis.
Requirements
Feature available from version 3.36.0
Only supported it New Client Experience
Categories are available exclusively with AWS Cloud Provider
The Quality Analysis add-on must be enabled in your XCALLY license to use this feature
Video Tutorial
Configuration
In the QA Categories section, you can view all created categories along with their associated cloud provider.
Each category is identified by a unique key, which combines the provider and the category name.
Select the appropriate AWS Region before creating or editing categories.
Clicking on Add category to create a new category:
And insert:
Unique Name: This serves as the category key. Spaces are not allowed—use underscores (
_) instead.AWS Region: The system automatically uses the region selected earlier, but you can change it if needed.
AWS Provider
Rules: Define rules for the category. You can add between 1 and 20 rules. The four supported rule types are described below.
Types of rules
You can configure 4 types of rules:
If you create multiple rules, all must be met to track the desired elements within the voice recording.
Interruption
This rule detects the presence or absence of interruptions in the Call Analytics transcript. It helps evaluate whether the agent interrupts the customer, the customer interrupts the agent, or neither.
Use cases:
Agent interrupts customer
Customer interrupts agent
Either participant interrupts the other (default option)
No interruptions (enable Negate option)
Applicable Fields | Description |
|---|---|
Negate | Enable to mark recordings without interruptions; disable to mark recordings with interruptions. |
Time Range | Absolute = Specify the time interval in milliseconds during which interruptions are detected. Relative = Specify the time interval as a percentage of the call duration. |
Role
| Choose whether to detect interruptions from the Agent, Customer, or Both (default). |
Threshold | Minimum duration (in milliseconds) of interruptions to be marked. |
Interval
| See detailed explanation in the paragraph below |
Non-Talk Time
This rule tracks the presence or absence of silent periods exceeding a set threshold during the call. While normal pauses are expected, longer silences may indicate holds or issues.
Use cases:
Presence of silence
Presence of speech (enable Negate option)
Applicable Fields | Description |
|---|---|
Negate | Enable to mark speech periods; disable to mark silence. |
Time Range | Absolute = Specify the time interval in milliseconds during which you wish to get a period of silence. Relative = Specify the time interval as a percentage of the call duration, during which you want to search for a period of silence |
Threshold | Minimum length of silence to mark (in milliseconds). |
Interval | See detailed explanation in the paragraph below |
Sentiment
This rule marks the presence or absence of specific sentiments detected in the transcript, applied by default to both participants.
Use cases:
Positive, Negative, Neutral, or Mixed sentiments present or absent during specific parts of the call.
Applicable Fields | Description |
|---|---|
Negate | Enable to mark sentiments not included in the request; disable to mark sentiments included. |
Time Range | Absolute = Specify the time interval in milliseconds during which you wish to search for the specified feelings. Relative = Specify the time interval as a percentage of the call duration, during which you want to search for the specified sentiments |
Role | Select sentiment types to mark: Positive, Negative, Neutral, or Mixed (mandatory). |
Interval | See detailed explanation in the paragraph below |
Transcript
This rule detects the presence or absence of specific words or phrases in the transcript. Multiple phrases can be separated by semicolons (;). The rule triggers if at least one phrase is found.
Use cases:
Custom words or phrases spoken by Agent, Customer, or both
Absence of specified words or phrases spoken by Agent, Customer, or both (enable Negate option)
Custom words or phrases that occur during a specific period
Applicable Fields | Description |
|---|---|
Negate | Enable to mark absence of specified phrases; disable to mark presence. |
Time Range | Absolute = specify the interval for searching phrases/words, in milliseconds Relative = specify the interval for searching phrases/words, in percentage |
Role
| Select Agent, Customer, or Both (default). |
Targets
| Specify exact words or phrases to detect (mandatory). Semantic variations are not recognized. For example, if the costumer says “talk to the manager” instead of “talk to a manager,” the content is not marked. |
Interval | See detailed explanation in the paragraph below |
Types of intervals
Start-End
First
Last
You can define time ranges using either Absolute or Relative intervals.
Absolute Time Range specify the interval in milliseconds:
Start - End (e.g. Start 1000 ms - End 12000 ms). Define a specific segment of the recording
First defines the interval from the beginning of the recording up to the specified time (e.g. Start of call to 12.000 ms)
Last defines the interval from a specified time to the end of the recording (e.g. from 13.000 to End of call)
Relative Time Range specify the interval as a percentage of the total call duration:
Start ( % ) - End ( % ) | Define a specific segment based on call percentage (e.g. 20% - 80%)
First: from the beginning of the recording up to a specific percentage (e.g. Start - 20%)
Last : from a specific percentage to the end of the recording (e.g. 80% - End)
Related topics