Answers

——This article helps you understand what Answers does, and how to use it to analyze the quality of each bot answer.

Entry: Data › Analysis › Answers.

Figure 1: Answers

Answers Overview

Before using Answers, familiarize yourself with its use case and purpose:

• Use case: Review data to determine whether a bot is operating abnormally—for example, whether its resource-based answer rate or flow invocation rate has decreased.
• Purpose: Quickly identify bot anomalies using detailed answer data filtered by time period, bot, business type, and channel.

How to Use Answers

Go to Data › Analysis › Answers.

• Global Filter Bar

• Time zone: GMT+8:00 Beijing by default.
• Conversation start time: The last 7 days by default, with a maximum range of 180 days.
• Bot: All bots by default. Multi-select and search are supported.
• Business type / Channel type / Consultation channel: All values by default. Consultation channel options depend on the selected channel type.

• Review Answer Quality

Answers is aggregated by date and includes answer counts, answer rates, and like/dislike data.

• The statistical dimension identifies the date associated with the data.

Type	Metric Details
Statistical dimension	Date

• Count metrics show the number of bot answers and invocations of each resource type.

Type	Metric Details
Count metrics	Total answers, answers using resources, answers without resources, direct answers, knowledge invocations, flow invocations, tool invocations, skill invocations

• Rate metrics show the percentage of each answer type and resource invocation type.

Type	Metric Details
Rate metrics	Resource-based answer rate, non-resource answer rate, direct answer rate, knowledge invocation rate, flow invocation rate, tool invocation rate, skill invocation rate

• Like/dislike metrics show how customers rated bot answers.

Type	Metric Details
Like/dislike metrics	Likes, dislikes

• Locate Problems Alongside Other Pages

1. Set the filter conditions and query the answer data for the selected period.
2. Focus on disliked or unresolved answers to identify quality issues.
3. Use Customer Questions to review high-frequency questions, then send the issue to AI Studio for optimization.

Note: The fields and metrics displayed on the page are subject to the actual product.