Make better product decisions with AI

Making good product decisions is hard.

You’ve got competing priorities, technical constraints, unclear feedback & more. Throw in natural human biases and irrationalities, and it's no wonder product decisions get messy. (The joys of being a PM.)

This is why PMs love a good prioritisation framework. They help make the decision process easier. These frameworks can be great, but we often use them as a mental crutch. We jump to a framework when we really should be taking the time to understand our decision, uncover biases and assumptions, and consider a broader range of context.

The good news is that AI is excellent at helping us do that. Not to make decisions for us, but to act as a thinking partner.

Luckily, I've built a system in ChatGPT to help you do just that. It's yours to steal:

The AI system for clearer product decisions

This system is completely free to use and I'm going to give you all of the instructions you need to set it up for yourself.

It's a 3-part system consisting of two custom GPTs and one project setup protocol. All built in ChatGPT.

There's a decision framing GPT, a bias & assumption checker GPT, and setup instructions for your personal product copilot.

This system will help you:

• Think about big product decisions clearly using the right mental models.

• Check your big product decisions for any hidden biases or assumptions.

• Set up an AI thinking partner who knows everything about your product.

Here's how the system works and how to use it:

1️⃣ Decision framing GPT

This is a custom GPT (chatbot) designed to help you think more clearly about any product situation you're faced with.

What it does:

• Engages in a two-way dialogue to understand the decision you're making.

• Helps you slow down, clarify your thinking, and approach the decision the right way.

• Uses decision theory to recommend the right mental models, frameworks, and ways to approach the situation.

How to use it:

1. Describe the product decision or situation you’re facing.

2. The GPT will ask follow-up questions to clarify the problem.

3. It will guide you through the right thinking process and suggest mental models that fit.

Access the decision framing GPT here: Product Decision Doctor

If you're interested in the prompt instructions used to create this custom GPT, you can see them on this Notion page.

This GPT is trained on foundational decision theory and uses a core set of mental models and frameworks to make recommendations. You can find them here (same Notion page).

2️⃣ Bias & assumption checker GPT

As humans, our decisions are full of biases and assumptions. This GPT helps you to uncover and address any assumptions or biases that you may have missed in your product decisions.

What it does:

• Reviews your decision for hidden biases and flawed assumptions.

• Highlights blind spots and helps you fix them.

How to use it:

1. Input your decision or idea.

2. The GPT will probe your reasoning and look for biases.

3. It will highlight issues and suggest ways to improve your thinking.

Access the bias checker GPT here: Bias & Assumption Checker for PMs

If you're interested in the prompt instructions used to create this custom GPT, you can see them on this Notion page.

This GPT is trained on over 40 core cognitive biases that are most likely to affect a product decision. You can find them here (same Notion page).

3️⃣ Your personal product copilot

What it does:

• Turns ChatGPT into a dedicated thinking partner for your product.

• Loads it with key context—your product strategy, roadmap, customer insights. This turns ChatGPT into an expert in your product so it can give smarter advice.

How to set it up:

1. Create a Project for Your Product Area

   • Use ChatGPT’s Projects feature to create a dedicated space for the product area you manage.

2. Give Clear Instructions

   • In the Instructions section, define ChatGPT’s role.

   • Use these instructions (Notion page). These are the exact instructions I use to create projects like this.

3. Add Context

   • This is the most important step. The more context you give ChatGPT, the more useful it becomes.

   • In the Files section of the project, add:

     • Company & product background

     • Strategy documents

     • Roadmaps

     • Customer feedback

     • Analytics and metrics

4. Ask Better Questions

   • Don’t ask: “What should I do?”

   • Instead, ask:

     • “Given our retention issues and these user interviews, what areas should we be focusing on?”

     • “We’ve been losing deals to [Competitor] due to [Sales Loss Reasons]. Based on what you know about our product and differentiation, what are some strategic angles we could consider?”

The key is to treat this like a living system. Keep the context and instructions fresh and relevant to get the best advice.

Using the System

This isn't a 1-2-3 step-by-step process. Treat this system like more of a loop. You might start with the decision framing GPT, dive into your product project for more context, and then run the idea through the bias checker. Or go back and forth until things click.

And remember this system is designed to set ChatGPT up as your thinking partner. But,you still need to do the critical thinking yourself. This should help you think more clearly, not do your thinking for you.

Product decisions are hard, but getting a big product decision right is one of the highest-leverage things a PM can do. Hopefully, this system can help you get more decisions right than wrong.

Try it out and let me know how it goes. I'd love to hear some feedback.