Conversation history

The conversation history tool provides an interface for browsing and analyzing actual production conversations between your agent and end-users. You can use this tool to evaluate the functional performance of your agent or to debug issues.

Flow and page analysis lets you understand traffic and escalations through the flow graph of an agent. This analysis lets you analyze traffic flow, discover issue hotspots (for example, to pinpoint high escalations), and drill down to the specific pages and intents causing issues. You can use flow and page analysis along with conversation history filters to study a relevant subset of traffic.

Limitations

Loading the flow analysis table and flow analysis graph views can be slow for agents with large configurations. To improve load times, use an intent filter and a date range of one week.

The system logs a maximum of 500 turns for each conversation.

Permissions required

Using this tool requires the dialogflow.conversations permission. You can add this permission to a custom role. It is also available in the administrator role and other Dialogflow CX roles.

Enable and disable conversation history

The conversation history tool displays conversations from your agent. To enable or disable this feature, go to the Enable conversation history setting.

You can also manage how long data is retained by using the retention_window_days security setting.

Audio playback

Conversation history provides audio playback of conversations. You can enable this feature in the general settings. If you don't enable this feature before you load a conversation, the console shows a notification that audio playback is unavailable.

This feature lets you listen to conversation audio directly in the console using signed URLs. Audio playback uses signed URLs to provide authenticated, time-limited access to audio files stored in your Cloud Storage buckets, removing the need to grant users direct access to Cloud Storage buckets.

To play audio, you must have permission to generate signed audio URLs. You must also have imported voice conversations into the console with audio URIs pointing to audio files in a Cloud Storage bucket.

Browse conversation history

To use the conversation history tool:

  1. Go to the Conversational Agents console.
  2. Select your project.
  3. Select your agent.
  4. In the navigation menu, select Conversation history.
  5. View the conversations in the table.
  6. Optionally apply filters using the filter control above the table.
  7. Select one of the tabs described later in this document.

Download conversation history to a CSV file

You can download a maximum of 50 conversations to a CSV file. Use the Filter conversations field on the Conversation history page to select the specific conversations you want to export. To export larger batches of conversations, use BigQuery.

  1. In the navigation menu, select Conversation history.
  2. Select a conversation session, then click Open conversation in preview panel.

    Open preview button

  3. Click Export all conversations. A CSV file containing the conversations downloads to your computer.

Conversations

You can view conversations in the Conversation summary view and the Conversation details view.

Conversation summary view

The Conversation summary view lists all conversations and provides the following metadata for each conversation:

X Item
Conversation ID An identifier for the conversation.
Duration The duration of the conversation.
Turns The number of conversational turns.
Channel Either chat (text), voice, or undetermined (no end-user input has been provided to the conversation yet).
Language The language for the conversation.
Environment The agent's environment.
Start time The start time of the conversation.
Flags Can be set to Live Agent Handoff flag (if conversation escalated to a human agent) or Abandoned (if end-user left the incomplete conversation) or TU,TD flag (if answer in the conversation got end-user rating).

You can filter the results by metadata (except environment and duration) and by the following additional filter options:

X Item
Intent The provided intent was matched at some point in the conversation.
No Match Some subset of conversational turns resulted in an intent no-match.
Flow The provided flow is the final active flow at the end of some conversation turn.
Page The provided page is the final active page at the end of some conversation turn.
Is Live Agent Handoff The Live Agent Handoff flag is set.
Is Abandoned The Abandoned flag is set.
Conversation ID A specific conversation is chosen.
Start Time A date range is provided.
Agent Utterance The provided text is a substring of an utterance that the agent used at some point in the conversation. The substring must contain complete words, bounded by space characters or either end of the string, as found in the utterance.
User Utterance The provided text is a substring of a user utterance. The substring must contain complete words, bounded by space characters or either end of the string, as found in the utterance.
Has Thumbs Up Feedback Thumbs Up feedback provided for a response in the conversation; the TU flag is set.
Has Thumbs Down Feedback Thumbs Down feedback provided for a response in the conversation; the TD flag is set.

Screenshot of conversation summary view

Conversation details view

You can browse the turn-by-turn view of a specific conversation in the Conversation details view. Each turn provides the end-user message, the agent message, and the following metadata:

X Item
Intent An intent display name that was matched or a no-match.
Page The final active page name.
Flow The final active flow name.
Play conversation Conversation playback control, including audio if available.
Parameters The parameters collected during the conversational turn.

Screenshot of conversation details view

Example use case

Identify agent issues that result in escalations to a human agent. From the Conversation summary view, apply the Is Live Agent Handoff filter for Yes.

Examine some of these conversations to find common patterns. For example, you might find that in most of the conversations, a No Match flag appears for turns on a specific page. This indicates that you need to create an intent route, or that an existing intent route is not in scope when it needs to be.

Flow analysis table

The Flow Analysis - Table tab lets you analyze traffic and exits in flows and pages in tabular form.

The following metrics are shown:

X Item
Flow name The flow name for each flow in your agent.
Relative traffic The number of conversations that went through this flow.
Escalation rate Percentage of conversations that result in a request for human escalation, out of all conversations that went through this flow.
Exit rate Percentage of conversations that ended after this flow or transitioned to SESSION_END from this flow, out of all conversations that went through this flow.

Screenshot of previously described interface

Click the list button to open a sample of relevant conversations for the row. You can read through the samples to better understand specific and nuanced issues. For example, you can examine why escalations are happening. To view the full details of a sample, click the Conversation details link.

Screenshot of previously described interface

Click Detailed Stats for a flow to view flow transition statistics for the selected flow.

The following metrics are shown:

X Item
Next flow Stats for transitions from selected flow to another flow.
Previous flow Stats for transitions from another flow to the selected flow.

Screenshot of previously described interface

Click any flow row to drill down into the pages that make up that flow.

The following metrics are shown:

X Item
Relative traffic The number of conversations that visited this page.
Escalation rate Percentage of conversations that result in a request for human escalation, out of all conversations that visited this page.
Exit rate Percentage of conversations that exit the flow or exit the session (END_SESSION) directly after visiting this page, out of all conversations that visited this page.
No match rate Percentage of conversations that resulted in a no match, out of all conversations that visited this page.

Screenshot of previously described interface

Click Detailed Stats for a page to view the traffic statistics, which show page transition stats from and to the selected page.

The following metrics are shown:

X Item
Intent Stats for intent matches for the page.
Next page Stats for transitions from selected page to another page.
Previous page Stats for transitions from another page to the selected page.

Screenshot of previously described interface

Flow analysis graph

The Flow Analysis Graph tab lets you analyze traffic and exits in flows and pages.

Click any flow to drill down into the traffic for the pages within that flow.

Screenshot of previously described interface

Use the zoom slider to vary the level of detail on the graph. Low volume paths are hidden from the visualization at lower zoom levels. Even at higher settings, very low volume paths are always hidden.

Screenshot of previously described interface

Metrics are color-coded to help you identify high severity issues.

Screenshot of previously described interface

Click the list button to show you relevant conversations.

Screenshot of previously described interface

Select a flow to view the pages within that flow. The following overlay metrics are available: relative traffic, no match rate, escalation rate, and exit rate.

Screenshot of previously described interface

Access conversation history with API

You can access conversation history with the API. See the V3beta1 reference documentation.

Privacy

When you enable conversation history, Google collects and stores your conversation data for a period of time before permanently deleting it. The default is 365 days, but you can shorten this time period using the retention_window_days field in SecuritySettings.

You retain full ownership of your data while it's stored. Google won't use your data or access it for any purposes other than customer support.