What are Outcomes?
Outcomes are what your end users receive at the end of your AI product — personalised, and built from what they told you.
An Outcome is the deliverable at the end of every AI product. Once the conversation is complete and the AI has collected everything it needs, it generates the Outcome — something specific to that person, built from what they shared.
This is what separates Productised from a lead form or a quiz. Your end user does not receive a generic results page. They receive something that reflects their situation.
The three Outcome types
Assessments
A fully custom HTML result page you design in Claude. Uses live tokens to pull in AI outputs, scores, and form answers. Best for branded, visual deliverables — scorecards, diagnostics, reports.
Advisor
After collection ends, the AI shifts into a live advisory session — speaking first with full context from the conversation, then opening the floor for follow-up questions.
Standard
The AI streams a personalised written output directly in the product — instant, readable, and built from what your end user shared. The fastest way to deliver a personalised result.
How Outcomes connect to your AI Product
Every Outcome is a node on the canvas. You drag it from the toolbar, drop it below your AI Product node, and draw a connection between them.
In most cases your AI Product connects first to an AI Agent node, which processes the collected data and produces the structured outputs the Outcome needs — then the AI Agent connects to the Outcome node.
See Building your canvas for the step-by-step.
What makes an Outcome personalised
The AI has the full conversation in context when it generates the Outcome — every answer, every nuance of what the person shared. It uses this to make the output specific.
A Standard output generated for a founder who said they're pre-revenue looks different from one generated for a founder with an established product. Not because you wrote two versions — because the AI understood the difference and responded to it.
This is the core of what Productised does: every Outcome is different because every conversation is different.
Your AI Agent assigns each user a named tier (e.g. "Growth-Ready", "Not Yet Ready") as it scores them. Give the agent guidance on how each tier's result should differ, and the Outcome adapts its content and CTA to that tier — no manual routing rules needed.