Stay organized with collections
Save and categorize content based on your preferences.
You can opt in to the Dialogflow runtime integration feature to integrate data
from other Dialogflow-powered Contact Center AI services into
Insights. These services also include
Agent Assist
and virtual agents.
Dialogflow runtime integration use cases
We recommend that you enable this feature if you use Agent Assist or Virtual
Agent to create conversations and you want to use
Insights features on
these conversations. For example, Dialogflow runtime integration is useful if
you want to:
View your conversations and their metadata from a user interface.
Run more machine learning analytics on your conversations.
Export your conversations to BigQuery for custom analysis or
visualization in Looker.
How Dialogflow runtime integration works
One-time sync for every new conversation
After you enable Dialogflow runtime integration, Insights will
create a representation of this
conversation
and its
associated metadata
each time you
complete a conversation
in Agent Assist or Virtual Agent. If the source conversation has a transcript,
Insights will generate
additional metadata
for it if you specified a percentage of conversations to be analyzed in your
analysis rule or project settings.
The default analysis percentage is 0.
If you modify the source data from Agent Assist or a Virtual Agent, the
Insights representation of the data won't be affected. The
opposite is also true.
Synced data can be queried and exported from Insights
Conversations exported to Insights will be queryable from
Insights along with their associated metadata. You can also
export conversation data to BigQuery
to perform your own raw queries.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[[["\u003cp\u003eDialogflow runtime integration allows you to integrate data from Dialogflow-powered services like Agent Assist and Virtual Agent into Insights.\u003c/p\u003e\n"],["\u003cp\u003eEnabling this feature lets you view and analyze conversations from Agent Assist or Virtual Agent using Insights features.\u003c/p\u003e\n"],["\u003cp\u003eInsights creates a representation of each new conversation and its metadata upon completion in Agent Assist or Virtual Agent.\u003c/p\u003e\n"],["\u003cp\u003eData synced to Insights can be queried and exported, including the option to export conversation data to BigQuery for custom analysis.\u003c/p\u003e\n"],["\u003cp\u003eInsights doesn't modify the source data from Agent Assist or Virtual agent, and in turn, modifications to the source data also won't affect Insights.\u003c/p\u003e\n"]]],[],null,["# Dialogflow runtime integration\n\nYou can opt in to the Dialogflow runtime integration feature to integrate data\nfrom other Dialogflow-powered Contact Center AI services into\nInsights. These services also include\n[Agent Assist](https://cloud.google.com/agent-assist/docs)\nand [virtual agents](/agent-assist/docs/agent-virtual).\n\nDialogflow runtime integration use cases\n----------------------------------------\n\nWe recommend that you enable this feature if you use Agent Assist or Virtual\nAgent to create conversations and you want to use\n[Insights features](/contact-center/insights/docs/how-to) on\nthese conversations. For example, Dialogflow runtime integration is useful if\nyou want to:\n\n- View your conversations and their metadata from a user interface.\n- Run more machine learning analytics on your conversations.\n- Export your conversations to BigQuery for custom analysis or visualization in Looker.\n\nHow Dialogflow runtime integration works\n----------------------------------------\n\n### One-time sync for every new conversation\n\nAfter you enable Dialogflow runtime integration, Insights will\ncreate a representation of this\n[conversation](/contact-center/insights/docs/reference/rpc/google.cloud.contactcenterinsights.v1#conversation)\nand its\n[associated metadata](/contact-center/insights/docs/reference/rpc/google.cloud.contactcenterinsights.v1#runtimeannotation)\neach time you\n[complete a conversation](/agent-assist/docs/smart-reply#create_a_conversation)\nin Agent Assist or Virtual Agent. If the source conversation has a transcript,\nInsights will generate\n[additional metadata](/contact-center/insights/docs/reference/rest/v1/projects.locations.conversations.analyses#callanalysismetadata)\nfor it if you specified a percentage of conversations to be analyzed in your\n[analysis rule](/contact-center/insights/docs/analysis-rule) or [project settings](/contact-center/insights/docs/reference/rest/v1/Settings).\nThe default analysis percentage is 0.\n\nIf you modify the source data from Agent Assist or a Virtual Agent, the\nInsights representation of the data won't be affected. The\nopposite is also true.\n\n### Synced data can be queried and exported from Insights\n\nConversations exported to Insights will be queryable from\nInsights along with their associated metadata. You can also\n[export conversation data to BigQuery](/contact-center/insights/docs/export)\nto perform your own raw queries.\n\nWhat's next\n-----------\n\nTo enable Dialogflow runtime integration, see the\n[Dialogflow runtime integration quickstart](/contact-center/insights/docs/enable-dialogflow-runtime-integration)."]]