Product design prompt generator Build an AI prompt for discovery, PRDs and your design workflow

This product design prompt generator turns a product problem into a structured AI prompt for ChatGPT, Claude, Gemini, Microsoft Copilot or Notion AI. It’s built for product designers, product managers and teams working with hypotheses, interviews and usage data. Founderz made it; our Certificate Program in Artificial Intelligence and Innovation teaches people to use AI at work.

Product design prompt generator

Describe the problem, idea or product decision and get a prompt that makes ChatGPT, Claude or Gemini help with discovery, PRDs, prioritization, metrics or launches without losing sight of the user and the evidence. Only step 1 is required: everything else stays on Auto.

How to use the product design prompt generator

Only the first step is required: anything you skip stays on Auto.

1. The essentials. Describe your problem in “Brief or context” and pick the type of work under “What do you need”, or switch on the option that makes the AI ask questions first.
2. Product and user. Set the “Product stage”, the “Primary user” and the “Business context”. Under “Advanced settings” you can add the user’s current alternative.
3. Evidence and decision. Pick the “Available evidence”, “How you will measure success” and the “Main constraint”. Under “Advanced settings” you set the trade-off, the scope and who decides.
4. Format and delivery. Pick the “Format”, “Where you will use it” and the “Answer language”. Under “Advanced settings” you get extra controls and a field for limits or available material.
5. Generate and copy your prompt. Click “Generate prompt”, enter your email to see the whole prompt and copy it.

Anything left on Auto stays out of the prompt, which keeps it a focused set of instructions. Fill in at least “Available evidence” and “How you will measure success”: those two fields separate an actionable product prompt from one that only produces fresh ideas.

Example: a prompt to improve onboarding for a survey tool

With the generator’s “Activation” example, the prompt starts like this. It’s an excerpt: the button copies the full prompt.

ROLE
Act as a senior product designer with discovery and delivery experience. You work with Jobs to be Done, the opportunity solution tree, the double diamond and RICE prioritization, and you always separate evidence from hypotheses: you never fill gaps with invented user data, metrics or results.

GOAL
Design the onboarding and activation improvement with an experiment to measure it.

CONTEXT
Product context provided by the user (treat it as your main source):
[START OF BRIEF]
Online survey tool. 60 % of new accounts do not create a single survey in their first week. We want to redesign the first few minutes so they send their first survey sooner.
[END OF BRIEF]

...

OUTPUT FORMAT
Deliver: the activation definition; a funnel table with columns Step, Users [to confirm], Drop off and Observed friction; the proposed changes with their hypothesis; and the experiment card.

...

The full prompt adds the guardrails against invented data, the seven working steps built around the activation metric and the final checks. It also keeps evidence and hypotheses clearly apart.

The result: the AI defines activation, locates the friction in the funnel and proposes changes with a measurable experiment, without inventing numbers.

Design tips for better AI-generated output

Common use cases in a design workflow

• A research and interview plan for discovery.
• A lightweight PRD with acceptance criteria for a new feature, ready to refine with engineering.
• A prioritized decision matrix for the roadmap.
• A metric analysis of activation or retention, or a review of support feedback by product category, with hypotheses and an analytics plan.

What makes good prompts to create product decisions

A good product prompt covers the role, the task, the business context, the material or evidence, the constraints and the output format. The difference from creative prompts is evidence: if you don’t tell generative AI tools what you know and what you only assume, they mix hypotheses with facts and hand you a proposal that looks validated when it isn’t. Ask the model to keep the two apart and to suggest a small experiment for each hypothesis.

Simple prompts with one clear goal beat long ones, and a focus on user experience, usability and ease of use beats a focus on layout alone. That habit is what makes AI design support useful for real product work.

Is this a design challenge generator?

No. Those tools give you a random exercise for graphic design practice, such as logo concepts, landing pages or a color palette and typography study, so graphic designers and non-designers can sharpen their skills and find creative inspiration.

This one works the other way round: you bring a problem from your product and it builds the prompt around it, following design principles instead of a random theme. If you want graphic design prompts to practice with, a practice tool fits better; if you have a product decision to make this week, use the form above.

More AI prompt templates for each use case

If you’d rather start from ready-made examples, the Founderz AI prompt library has prompts to simplify specific tasks across product, design and other teams.

• Interface prompts To generate design briefs for screens, flows and states.
• Prompts for project managers For plans, risks and status reports.
• Prompts for branding For positioning and brand identity.

Why I built these generators

Frequently asked questions

Is the product design prompt generator free?

Yes, it’s free. Before it asks for anything, it shows you a preview with the first sections of the prompt. To see the whole prompt and copy it, enter your email and accept the Privacy Policy; receiving communications from Founderz is optional.

It only asks once per visit: any prompts you generate after that are shown in full.

Can it write a PRD?

It writes the prompt for one. Choose “PRD and acceptance criteria” under “What do you need” and adjust the “Format” if you want tables or backlog tickets. The assistant drafts it; review the scope and criteria with engineering before you treat it as final.

Can I paste user data?

Only aggregated or anonymized data. The form doesn’t send your brief to any model, but the assistant you paste it into will process it, so follow your company’s data policy.

How is it different from the interface prompts?

This one handles the product decision: problem, evidence, metric and prioritization. The interface one handles screens, states, components and microcopy. You can use them in sequence.

What language does the prompt come out in?

The one you pick under “Answer language”: English, Spanish or English and Spanish. If you leave it on Auto, the answer comes back in the language of your brief.

Last updated: October 26, 2026