The Product CanvasUsing the AI assistant

Using the AI assistant

The AI assistant is your build partner on the canvas — ask it to improve, adjust, or extend your product while you work.

While you're building on the canvas, there's an AI assistant available to help you. It's the chat panel that opens from the sparkle button on the left of the floating toolbar. This is your build-time collaborator — a way to make changes, get suggestions, and improve your product through conversation rather than manual configuration.

The AI assistant is a tool for you as the builder, while you're designing your product. It's not the same as the conversational AI that your audience experiences. Keep reading to understand the difference.

Opening the AI assistant

Click the sparkle (bolt) button on the far left of the floating toolbar at the bottom of the canvas. The AI assistant panel slides in from the side. You can keep it open while you work on the canvas — changes it makes are reflected on your nodes in real time.

📸 Screenshot needed: The canvas with the AI assistant panel open on the right side — showing a conversation thread with a suggested change highlighted and the AI's explanation visible

What you can ask it

The AI assistant understands your product in full — it knows your current configuration, your questions, your decision logic, and your outcome type. You can have a natural conversation with it about any aspect of your product.

Here are the kinds of things you can ask:

Adjusting questions and conversation flow:

  • "Make the conversation feel less formal — my audience is small business owners, not executives"
  • "Add a question about their current revenue before the goals question"
  • "Remove the phone number question and replace it with a question about their biggest challenge"

Changing collection behaviour:

  • "Switch this to adaptive mode so the AI asks follow-up questions based on what they share"
  • "Add a scoring model so high-scoring leads automatically get a different outcome"
  • "Make the readiness question required and everything else optional"

Changing or configuring outcomes:

  • "Change the outcome to a personalised document instead of a report"
  • "Add a call to action at the end pointing to my calendar link"
  • "Make the report style more executive — fewer charts, more narrative"

Adding Powers:

  • "Add a Knowledge power so the AI can reference my methodology document"
  • "Add web research so the AI can pull in current data about the person's industry"

Troubleshooting and improving:

  • "Check if the collection goals make sense for a lead qualification tool"
  • "The conversation feels too long — suggest which questions to cut"
  • "What outcome type would work best for a business diagnostic?"

How the AI assistant makes changes

When you give the AI assistant an instruction, it updates your product configuration directly. You'll see the changes reflected on the canvas and in the settings panel immediately. The assistant explains what it changed and why, so you're always in control.

If you don't like a change, you can tell it to revert, or simply undo using the standard keyboard shortcut.

The AI assistant is most powerful when you're specific. "Make this better" gives it less to work with than "Make the opening question more direct — my audience is time-poor and values brevity."

The difference between the AI assistant and your AI Product

This is worth understanding clearly:

The AI assistant helps you build the AI Product. They're two separate things operating at two different levels.