Analytics tags
Define custom tags in your workspace to track performance of conversations in your agentic CX designer workspace
Analytics tags
Analytics tags help you mark important moments in a conversation flow so you can track how users move through your agentic CX designer application.
A tag can be applied to a specific node in a flow. When a user reaches that node during a deployed conversation, the tag is recorded. You can then use tags to review flow traversal, filter conversation history, or create analytics dashboard charts around milestones that matter to your team.
Use analytics tags to track moments such as:
Task started or completed
User reached fallback or escalation
Booking completed
Offer accepted or declined
Knowledge base answer shown
Payment handoff started
User reached a key decision point
Analytics tags do not automatically determine whether a conversation was successful, helpful, or automated. They mark the points you choose so you can measure and review them later.
To access analytics tags, open your workspace menu, select Resources, then choose Analytics tags.
Why analytics tags matter
Users may not always complete a conversation in the way you expect. They may abandon a flow, loop through a retry path, escalate to a human agent, or complete a task successfully.
By placing analytics tags at strategic points in a flow, you can better understand where users are going and what outcomes they are reaching.
For example:
booking_started
First node in a booking flow
How many users begin the booking process.
booking_completed
Confirmation node
How many users complete the booking.
fallback_reached
Fallback or recovery path
How often users hit an error or unsupported path.
escalation_requested
Human handoff node
How often users request or require escalation.
offer_declined
Decline path after an offer
How often users reject an offer.
These tags can help you identify completion rates, drop-off points, repeated issues, and opportunities to improve the experience.
System tags
Agentic CX designer includes default system tags that can be used in flows:
automated
helpful
not_helpful
These tags are available by default, but they do not automatically classify a conversation. You decide where to place them based on what each point in your flow represents.
For example, you might place automated on a node reached after the application completes a task without human handoff, or not_helpful on a path where the user indicates the answer did not solve their issue.
Custom analytics tags
You can create custom analytics tags for the milestones, outcomes, or events that matter to your application.
To add a custom analytics tag:
Open Resources.
Select Analytics tags.
Choose Add tag.
Enter the tag details.
Save the tag.
Custom tag fields include:
Label
The name of the tag that appears when tagging nodes in a flow. Use a clear, descriptive name.
Type
Categorizes the tag for analytics reporting, if available.
Description
Explains what the tag is intended to track. This description is visible from the Analytics tags resource page.
Tag labels should be easy to understand and consistent across the workspace. Use names that describe the event clearly, such as booking_completed, fallback_reached, or agent_handoff_started.
Adding analytics tags to nodes
After a tag is created, you can apply it to nodes in a flow.
To add an analytics tag to a node:
Open the flow in the Canvas.
Select the node you want to track.
Open the node’s Functionality menu.
Choose Analytics tags.
Select one or more tags.
Save the flow.
Once the updated flow is included in a deployed build, future conversations that reach the tagged node will record the tag.
Viewing tag activity
You can review tag activity in several places:
In-Canvas analytics
Filter the flow view to conversations that reached tagged nodes.
Conversation transcripts
Filter transcripts by selected tags to review sessions where a milestone occurred.
Analytics dashboards
Create dashboard charts that track tag activity over time.
Use tags when you want to move beyond general conversation volume and monitor specific outcomes or checkpoints inside the experience.
Deleting a tag
To delete a custom tag:
Open Resources.
Select Analytics tags.
Expand the tag details.
Choose Delete tag.
Before deleting a tag, confirm that it is no longer needed for active reporting or flow analysis.
Important notes
Analytics tags only apply to conversations that occur after the tag is added to a node and deployed.
Tags are not applied retroactively to previous conversations.
A node can have more than one tag, but using one tag per node is often easier to interpret and helps avoid confusing counts.
Tags mark that a node was reached; they do not automatically explain why the user reached that node.
Use clear tag names and descriptions so teammates understand what each tag is intended to measure.
Last updated

