AI Graph Explainer

The AI Graph Explainer is a conversational assistant designed to help users interpret complex graphs, debug configurations, and understand journey behaviors more intuitively. Powered by Xponent Intelligence, it uses Generative AI and documentation content for Xponent Journey Hub to deliver plain-language explanations in context. It enables users to:

Ask natural-language questions about a selected graph. You can ask the Graph Explainer a range of questions, including but not limited to:
- What does this graph do?
- What schema variables are being referenced in this graph?
- Why is this graph invalid?
- How many "Journey steps" were updated in this graph?
  - Follow-up question - can you list journey steps in order as it is being updated?
Get help understanding node connections and logic.
Troubleshoot or debug invalid graphs.
Receive detailed responses based on the quality of the prompt/question.

The AI Graph Explainer is powered by Xponent Intelligence. To use this feature, Xponent Intelligence must be enabled for your organization.

To access the AI Graph Explainer, hover over the icon located at the bottom-right corner of the screen. The icon will expand slightly to indicate it is interactive. When the explainer window is in its collapsed state, you can drag it up or down to adjust its position, making it easier to use the left-side schema panel without obstruction. Click on it to open the full explainer window and start interacting with the assistant.

The fully opened AI Graph Explainer window can be dragged horizontally across the screen as needed. Click and hold the drag icon on the top left of the panel, move it left or right, and release to place it where it’s most convenient for you.

When working with complex or extensive journey graphs, the AI Graph Explainer may not be able to process them if they exceed the supported size threshold. If the selected graph is too large to process, the interface displays a prompt informing the user that the graph size exceeds the allowable limit.

Error Handling Prompt Includes:

A message advising users to simplify the graph before trying again.
A suggestion to contact support if the issue persists and the graph cannot be reduced in size.

The quality of your question affects the quality of the response. Try to ask specific and clear questions to get more detailed and helpful responses.

User Actions in the AI Graph Explainer Interface

The AI Graph Explainer interface is designed for intuitive interaction. It provides key actions that enhance user control, enable feedback, and support session management. After the assistant generates a response, the AI Disclaimer and feedback response buttons appear at the end of the conversation.

AI Disclaimer

The disclaimer informs users that they are interacting with an AI Graph Explainer. It serves to set expectations and promote responsible AI usage.

You are interacting with a Generative AI assistant. This assistant uses AI to provide responses based on patterns in data. While it can offer helpful information and support, it is important to keep in mind:

Responses may not always be accurate or complete.
The assistant has no real-time awareness or personal knowledge of the user.
It must not be relied on for legal, medical, or safety-critical decisions.

This disclaimer helps ensure responsible AI use and sets clear expectations for users.

Clear Chat

Click on the Clear Chat button located at the top of the AI Graph Explainer interface. When clicked, a confirmation prompt appears to inform that the current conversation will be permanently deleted and cannot be recovered. To proceed, click Yes, delete.

This action resets the current conversation with the Graph Explainer.

Use this when switching to a different graph or starting a new inquiry.
Clearing the chat ensures the assistant no longer retains previous context and starts fresh.

Feedback Response Button

After receiving a response from the AI Graph Explainer, users have the option to provide conversation feedback on the usefulness of the answer by selecting a 👍 (Good) or 👎 (Not Good) icon. This helps the product team continuously improve the AI performance and user experience.

👍 Good: Click on the thumbs up icon if you found the AI response helpful or accurate. A confirmation message will appear: “Thank you for your feedback.” No additional action is required.
👎 Not Good: Click on the thumbs down icon if the response was not helpful, incomplete, or inaccurate. You will be prompted to provide more context to help the team understand and address the issue:
- Select a feedback category from the provided list.
- Describe the issue using the free-text field. You can include what was missing, unclear, or incorrect in the AI’s response.
- Click Submit to send the extended feedback.

This feedback is reviewed by the Xponent Intelligence team to refine AI performance and improve the quality of future responses.

Feedback

Clicking Feedback opens a Microsoft Form where users can report issues or share improvement suggestions. The form includes structured questions to help the product team gather context-rich feedback.

Select a category that best classifies your feedback.
Describe the issue (include as much context as possible, including the steps taken).
Have you encountered this issue more than once? (Yes/No). If yes, then how often? This helps assess severity and frequency.
What was the expected behavior or outcome?
For Example: I expected the Explainer to provide a breakdown of the journey logic.
Attach supporting files: Screenshots or logs.