Spotlight highlights insights and helps you handpick critical conversations for your QA reviews. This improves and speeds up your evaluation process by automatically surfacing the most valuable Improvement and learning opportunities.
You can learn more in this article or by watching this overview video.
This article contains the following topics:
- Understanding Spotlight insights
- Spotlight insights types
- Customizing voice call Spotlight insights
- Creating new Spotlight insights
Related articles
Understanding Spotlight insights
Spotlight is Zendesk QA’s AI-powered conversation discovery tool. It analyzes 100% of your team's interactions and includes a variety of out-of-the-box filters to identify must-review conversations.
Spotlight insight filters are available in both the Conversation view and in Assignments.
Spotlight insight types
You can use Spotlight to find conversations that include the following insight types:
Outliers spotlight insight
The Outliers insight offers the highest learning potential by allowing you to find critical conversations for review with a single click. It automatically identifies must-review interactions that are atypical or unusual for the team.
This filter flags conversations where:
- The customer issue was more complicated.
- The customer's problem could not be explained in one message.
- The agent needed to ask for further information to understand the request.
- There were more replies than average from both the customer and agent.
- The length of replies was longer than average.
Churn risk spotlight insight
The Churn risk insight highlights conversations where customers express a potential risk of attrition. It identifies instances where customers explicitly mention canceling their subscription or switching to a competitor.
Escalation spotlight insight
The Escalation insight flags conversations where the customer requests to speak with a higher-level representative, such as a manager. It does not detect internal escalation processes that occur outside the conversation.
Follow up spotlight insight
The Follow up insight flags instances where a support representative has promised to take a future course of action. It does not assess the validity of the action, only whether it was completed.
Exceptional service spotlight insight
The Exceptional service insight identifies instances where a support representative provided exceptional service and the customer expressed gratitude.
Sentiment spotlight insight
Understand how your customers feel when interacting with your support team through sentiment analysis. The Sentiment insight detects both negative and positive sentiments in conversations, allowing you to identify dissatisfaction or delight and address critical issues. See Understanding and applying the Sentiment filter.
Bot communication efficiency spotlight insight
The Bot communication efficiency insight compares your bot’s conversation handling to that of average agents. It returns an efficiency percentage that indicates whether interacting with the bot resolved the issue faster and with fewer questions than speaking with a human. Efficiency percentages under 20% are not returned.
Bot repetition spotlight insight
The Bot repetition insight reports when the bot is stuck in a loop and repeats the same message to the customer. Filter values include ‘detected’ and ‘not detected’.
Voice interactions spotlight insights: dead air and recording disclosure missing
Voice interactions spotlight insights detect dead air and missing call recording disclosures.
Dead air highlights calls where the gap between consecutive messages exceeds the set threshold. The default industry threshold is 30 seconds, but it can be adjusted to any duration.
Missing call recording disclosure detects whether the speaker discloses that the conversation is being recorded.
Customizing voice call Spotlight insights
You can customize the following existing voice call AutoQA Spotlight insights: Dead air and Recording disclosure missing.
To customize your voice call Spotlight insights
- In Zendesk QA, click your profile icon in the bottom-left corner.
- Click the Settings (
) icon. - Under Account in the sidebar, click AutoQA.
If necessary, click the Sidebar toggle icon (
) to display the side menu. - Select Insights.
- Click one of the following Spotlight insights to open and edit:
- Dead air: This insight analyzes moments of dead air in conversations lasting longer than the set threshold.
- Recording disclosure missing: This insight automatically identifies calls lacking the mandatory disclosure statement, such as 'This call will be recorded'. It uses a Large Language Model (LLM) that dynamically understands context, eliminating the need for preset phrases.
- Select whether the rule applies to all or specific conversations.
For specific conversations, choose the conversation source type, channel, helpdesk tag, and call direction (inbound or outbound).
- For the Dead air insight filter,define the maximum dead air threshold for your calls. For example, 30 seconds, the industry standard.
- Click Apply filter.
- Click Create Spotlight filter.
Creating new Spotlight insights
Newly created Spotlight insights work on text-based conversations and voice transcripts, automatically surfacing closed conversations by identifying specific keywords or phrases.
You can create up to 20 new Spotlight insights.
To create a new Spotlight insights
- In Zendesk QA, click your profile icon in the bottom-left corner.
- Click the Settings () icon.
- Under Account in the sidebar, click AutoQA.
If necessary, click the Sidebar toggle icon (
) to display the side menu. - Select Insights.
- Click Create Spotlight insight.
- Enter a name for your new Spotlight insight.
- Select the Insight type: negative, neutral, or positive.
- Select whether it applies to All conversations or Specific conversations only. Selecting Specific conversations allows you to narrow down where your rule applies by source type, conversation channel, help desk tag, and call directions (inbound or outbound).
- Click Agent message and specify whether the content applies to the agent, client, or any message, and if it should be an exact match or contain the specified words or phrases.
- Type or paste your text content, with one word or sentence per row.
You can add multiple rows and use different conditions for complex findings: and, or, is/not, contains/not any of, is/not any of, and variables. To use dynamic content type “/”. You can then use {··} to exclude names, company details, or other sensitive information during AutoQA's checks. For example, instead of "Take care, John" you can use "Take care, {··}" to represent any name.
- Click Apply filter.
- Click Create spotlight insight.