Verified AI summary ◀▼
Rating categories for scorecards help you evaluate customer conversations by focusing on key areas like spelling, comprehension, and tone. You can create custom categories tailored to your needs, including manual and AI-based options. Automatic scoring analyzes new conversations, providing insights to improve support quality. Manual categories are scored by human reviewers, offering flexibility in assessing interactions.
In Zendesk Quality assurance (QA), while spotlight insights flag tickets, rating categories score agents.
Well-defined rating categories streamline the review process by evaluating conversations with customers and ensuring consistent quality assessments in Zendesk QA. You can derive the Internal Quality Score (IQS) from rating categories, providing valuable insights to assess and improve your customer support.
There are many ways to combine rating categories to build your scorecard rating system. It may take some trial and error to identify the right factors to rate for accurately assessing and improving your customer support. Zendesk QA includes several system categories that are scored automatically when active. The following rating categories are among the most commonly used in Zendesk QA and provide a good starting point:
- Spelling and grammar: Detects and highlights grammar mistakes, misspellings, and style errors.
- Comprehension: Checks whether the agent understood the customer's issue, possibly requiring clarifying questions or summarizing the problem.
- Solution offered: Identifies whether the agent proposed a solution during the conversation.
- Tone of voice: Recognizes tones and calculates a score based on their positive or negative weights.
- Closing: Evaluates whether the agent closed the conversation properly, including offering further help and thanking the customer.
You can customize some of these categories and create custom rating categories tailored to your organization’s specific needs. Specifically, you can also:
- Create AI prompt-based categories
- Create exact text-match custom rating categories
- Create manual custom rating categories
Before creating categories, it’s important to understand the following information about categories and how they’re used in Zendesk QA:
- Enabling automatic scoring with AutoQA allows Zendesk QA to automatically evaluate and score agents.
- System autoscoring and AI prompt-based categories are indicated by a hologram
icon (
). See Viewing and managing rating categories for scorecards. - To ensure proper functionality for prompt-based AI insights, both the Automatic scoring with AutoQA and the LLM-based AutoQA account settings must be turned on.
- When a new category is created, only new incoming conversations are automatically analyzed and rated based on the category conditions. It may take some time for the scores to load due to necessary calculations.
- To qualify for analysis, conversations must include at least one message from both the customer and the agent, with a minimum word count of 10.
- AutoQA does not evaluate existing conversations, spam conversations, or historical data. However, existing conversations with a status of Solved or Closed that are updated after the category is added are also automatically analyzed and rated. As a result, you may notice a conversation that has already been analyzed and rated being analyzed and rated again in a new workspace.
- If no matches are found, "N/A" is assigned. AutoQA may also return "N/A" for certain categories when there is insufficient information to provide an accurate rating.