Verified AI summary ◀▼
Autoscoring evaluates and scores all customer interactions automatically, helping you maintain consistent quality assessments and save time. By activating autoscoring categories, you can focus on areas needing attention. Customize or create new categories to suit your needs. Use the AutoQA dashboard to track performance trends and identify training needs across categories like empathy, tone, and comprehension.
Autoscoring supports your manual review efforts by automatically evaluating and scoring customer interactions for 100% of your ticket volume based on predefined categories. This ensures consistent quality assessments, reduces subjectivity, and saves reviewers time, allowing them to focus on categories that need more attention.
This article contains the following topics:
Related articles
Overview of autoscoring (video)
In addition to the details in this article, this video provides a helpful visual overview of autoscoring.
About autoscoring categories
Autoscoring categories are indicated by a hologram icon (
). To use Automatic scoring with AutoQA,
you must first activate this feature in your account settings. Additionally, categories must be
marked as active and assigned to at least one workspace and one active
scorecard to automatically evaluate and score those agents.
Zendesk QA includes several system categories that are automatically scored when active. Text-based interactions (chats and emails) are reviewed by AutoQA separately from voice calls.
Admins can customize some of these categories and create new custom autoscoring categories.
The languages supported for each category depend on whether LLM-based AutoQA is activated in the Account settings.
The following system categories receive automatic scoring in Zendesk QA:
| Category | Description | Score | Languages/OpenAI-enabled | Languages/OpenAI-disabled |
| Greeting | Answers the question “Did the agent greet the customer?”. Greeting phrases in different languages are searched in agent messages. Each agent receives their own greeting score. If the first agent in the conversation uses a greeting, all agents receive an upvote. If a greeting is not found, the first agent receives a downvote and all other agents are scored as N/A. The Greeting AutoQA category is available for both text and voice interactions. |
Binary: ( ,  ) |
All | English, German, Spanish, French, Italian, Dutch, Polish, Portuguese |
| Empathy |
Assesses whether the agent was empathetic towards the customer and their problems. Each agent receives their own empathy score. However, the conversation empathy category is evaluated positively only if all agents demonstrate empathy. Empathy is evaluated negatively if at least one agent doesn’t show empathy. If, for some reason, such as the ticket being too short, the system fails to evaluate empathy, all agents receive a score of N/A. The Empathy AutoQA category is available for both text and voice interactions. |
Binary: ( , ) |
All | Not available without OpenAI |
| Spelling and grammar | Detects grammar mistakes, misspellings, and style errors. Error highlighting can be toggled on or off. Hovering over the highlight displays specific information about the error. Each agent receives their own spelling and grammar score. |
Customizable | English (US),English (UK), German, French, Polish, Spanish, Portuguese (Brazil), Portuguese (Portugal), Dutch | English (US),English (UK), German, French, Polish, Spanish, Portuguese (Brazil), Portuguese (Portugal), Dutch |
| Closing | Evaluates whether the agent closed the conversation properly,
including offering further help and thanking the customer. The conversation closing category is evaluated positively if at least one agent uses a closing. The Closing AutoQA category is available for both text and voice interactions. |
Binary: ( , ) |
All | English, German, Spanish, French, Italian, Dutch, Polish, Portuguese |
| Solution offered |
Identifies whether a solution was proposed in the conversation. The conversation solution category is evaluated positively if at least one agent provides a solution. The Solution AutoQA category is available for both text and voice interactions. |
( , ), or "N/A" if not sure |
All | Not available without OpenAI |
| Tone |
Recognizes tones and calculates a score based on their positive or negative weights. The conversation tone category is evaluated positively if at least one agent uses a positive tone. There are seven root categories in total, tied to tones or emotions:
|
Customizable | All | Not available without OpenAI |
| Readability | Analyzes how easily a text can be understood, considering word
complexity and sentence length. Complex words and long sentences highlighting can be toggled on or off. Each agent receives their own spelling and grammar score. |
Fixed 3-point scale | English, Spanish | English, Spanish |
| Comprehension | Checks whether the agent understood the customer's issue, possibly
requiring clarifying questions or summarizing the problem. The conversation comprehension category is evaluated positively if at least one agent is assessed as having understood the customer’s issue. The Comprehension AutoQA category is available for both text and voice interactions. |
Binary: ( , ) |
All | Not available without OpenAI |
Viewing autoscoring in conversations
Conversations with autoscoring categories are indicated by a hologram icon (
) in the Conversations view.
To access AutoQA in conversations
- In Quality assurance, click
Conversations in the sidebar. - Select a conversation.You can use custom filters to find conversations using AutoQA categories to review.
- In the sidebar on the right, click the Feedback
icon (
). - View the AutoQA autoscores under Reviews.
