The AutoQA Dashboard provides insights into both automatically and manually reviewed conversations, as well as agent data, and other performance metrics. This helps customer support teams evaluate and enhance their service quality.

This article contains the following sections:

  • Understanding the AutoQA Dashboard
  • Applying AutoQA Dashboard Filters
  • Drilling down into AutoQA Dashboard data
  • FAQ

Using the AutoQA Dashboard

To use the AutoQA Dashboard

  1. Navigate to Dashboards > AutoQA in Zendesk QA.
  2. Review the various metrics and insights provided (explained in detail below).
  3. Use the filters to narrow down the data based on specific criteria.
  4. Analyze the data to identify areas for improvement and track performance over time.

Understanding the AutoQA Dashboard

The AutoQA Dashboard provides information broken down into 17 categories.

  • The Auto-reviewed conversations metric shows the number of conversations that had at least one category automatically scored, without human intervention.

  • The Manually reviewed conversations metric shows the number of conversations that have been manually reviewed by humans.

  • The Efficiency gain metric displays how many times more AutoQA can cover tickets compared with manual reviewing by a human.

  • The Auto-reviewed per reviewee metric shows the average number of reviews by AutoQA for each reviewee.

  • The Manually reviewed per reviewee metric gives the average number of reviews manually given to each reviewee.

  • The Acceptance rate is the percentage of AutoQA scores that are accepted by users as accurate.

  • The Modified AutoQA conversations metric shows the number of conversations where a human has changed the AutoQA score in at least one category.

  • The AQS metric shows the average Agent Quality Score of all conversations that were evaluated by AutoQA.

  • The AQS over time metric reflects the difference of the Agent Quality Score over a selected period of time.

  • The Breakdown for top languages graph shows the ratio and count of conversation languages.

     

  • The Category Insights table provides a breakdown of AutoQA scores per categories.

  • The Category Scores over Time line graph shows changes in category scores over a selected period of time.

  • The AutoQA Conversations by Category graph provides the number of conversations that got an AutoQA rating (provided the conversations were closed with at least 1 public reply by the agent and customer).

  • The Category Scores per Reviewee table shows information on reviewees and their automatically calculated average score of specific auto category for selected time period.

  • The Root causes of spelling and grammar table gives information about reviewees and the AutoQA-assessed root causes picked up by AutoQA's Spelling and grammar category.

  • The Root causes of tone table gives information about reviewees and the AutoQA-assessed root causes picked up by AutoQA's Tone category.

  • The Repeated spelling and grammar mistakes per reviewee table lists repeated errors, broken down by reviewee and frequency
  • Last Data Update Time (UTC) provides a date- and timestamp of when the last dashboard update occurred. The AutoQA Dashboard is updated every hour. Updates are automatic and cannot be "forced". ​​

Applying AutoQA Dashboard filters

Data is sorted by conversation creation date. You can apply filters to drill into the data. You can toggle to Show/Hide filters as needed.  

  • The Conversation created date filter displays results based on the selected time frame (conversation creation date).
  • The Workspace filter displays results based on the selected workspaces that have any AutoQA review results.
  • The Root cause filter displays results based on available root causes, broken down by Grammar, Spelling and Style.
  • The Reviewee filter displays results based on the selected reviewees.
  • The Language filter displays results based on the selected conversation languages.
  • The Rating category filter displays results based on the selected auto categories.
  • The Sentiment filter displays results based on customer sentiment.
  • The Conversation channel filter displays results based on set channels.
  • The Connections filter displays results based on set connections.
  • The Outliers filter displays results based on your selection of Yes or No for Outliers.
  • The Rating score filter displays results based on the matches for the values you select.

Drilling down into AutoQA Dashboard data

You can drill into the following data points:

  • Auto-reviewed conversations
  • Manually reviewed conversations
  • Acceptance rate
  • Modified AutoQA conversation cards
  • Category Insights
  • Category Scores per Reviewee
  • Root causes of spelling and grammar.

To drill into the data

  1. From the AutoQA Dashboard, click on the metric you want to drill down into.
  2. From the dropdown, choose Drill into conversations.
  3. From the drilled-down screen, you can choose to Download the data.

FAQ

  • Why am I seeing more than one AutoQA category in the filter?
    Currently archived and deleted categories are also shown in the category filter.
  • If I manually change a rating that was auto scored as N/A does that affect the acceptance rate?
    Acceptance rate takes into account any changes to given auto score when submitting, including for N/A.
Powered by Zendesk