Flow traffic

Quickly analyze usage of your conversation flow

Viewing traffic with in-Canvas analytics

In-Canvas analytics helps you understand how users move through a deployed flow directly from the Canvas.

When analytics mode is enabled, each node displays traffic data showing how many unique conversations reached that node. This helps you identify common paths, underused branches, drop-off points, A/B testing results, and whether users are reaching the intended completion steps.

Use In-Canvas analytics to answer questions like:

• Which nodes are users reaching most often?

• Where are users dropping off?

• Are users reaching the expected success path?

• Are fallback or escalation paths being used more than expected?

• Did a specific conversation follow the path you expected?

• Does the flow behave as expected after deployment?

Important: A build that includes the flow must be deployed before In-Canvas analytics can display conversation traffic.

---

How to access In-Canvas analytics

To view In-Canvas analytics:

1. Open Resources from the workspace menu.

2. Select Flows.

3. Choose the flow you want to review.

4. Select the Analytics icon from the Canvas toolbar.

5. Review the traffic indicators displayed on each node.

To turn off In-Canvas analytics, select the Analytics icon again.

You can continue editing a flow while analytics mode is enabled. If you make changes that should affect live behavior, create a new build and deploy it before expecting updated deployed traffic to reflect the new flow version.

---

What the node indicators show

When In-Canvas analytics is enabled, each node displays a number that represents how many unique conversations reached that node during the selected filter period.

Use these indicators to quickly compare:

Signal	What it may indicate
High traffic on a node	Many users are reaching this point in the flow.
Low traffic on an expected path	Users may not be reaching the intended route.
High traffic on fallback or escalation paths	Users may be confused, unsupported, or encountering errors.
Drop-off before a completion node	The flow may be too long, unclear, or blocked by missing information.
Unexpected traffic on a branch	Routing, conditions, or state may need review.

In-Canvas analytics is especially useful when you want to connect a performance signal to the exact path users took inside a flow.

---

Filters

Use filters to refine the analytics view and focus on the conversations that matter most.

Filter	Use
Application	Review data for a specific application, especially when a flow is attached to more than one application.
Conversation ID	View the path taken during one specific conversation retreived from conversation history.
Start time	Set the beginning of the date and time range.
End time	Set the end of the date and time range.
Tag	Focus on conversations that reached nodes with selected analytics tags.

Filters help you narrow your review when investigating a release window, a specific user session, a known issue, or a tagged milestone.

---

Using Conversation ID

You can use a Conversation ID to inspect the exact path from one conversation:

1. Open the deployed application.

2. Go to Observe > History.

3. Open a conversation from the Transcripts table.

4. Copy the Conversation ID.

5. Open the flow involved in that conversation.

6. Enable In-Canvas analytics.

7. Paste the Conversation ID into the filter field.

8. Review the path the user took through the flow.

This is helpful when a transcript shows unexpected fallback behavior, repeated questions, missed routing, a failed integration, or user drop-off.