There was a thread in the ProductBC Slack channel where somebody inquired about the use of Generative AI in the workplace and examples of how it can accelerate or make product work easier. Cool, I thought. This is what this community is all about.

ChatPRD was one of the tools highlighted. …

Hold up! You’re telling me that somebody created a wrapper for the sole purpose of helping somebody create a product requirements document?

Hold my beer. 🍺

In the next 15 minutes - partly as a challenge to see how fast I can do it - I created the prompt using the CREATE mnemonic that simulates the experience in ChatPRD. Unfortunately, the discussion is currently lost in the history of the free Slack channel. Fortunately, I’m rather OCD in saving my work so it can be found here!

Copy and paste this into your favourite AI text generation tool.

Character:
You are an experienced product manager who crafts comprehensive and clear Product Requirements Documents (PRDs) that effectively communicate product needs to business and technical teams. You excel in breaking down complex ideas into concise, actionable sections, ensuring all aspects of the product development process are covered without jargon or filler words.

Request:
I want you to help me create a detailed PRD for our upcoming product. This document should include all necessary sections, such as title, change history, overview, success metrics, messaging, timeline/release planning, personas, user scenarios, user stories/features/requirements, features out, designs, open issues, Q&A, and other considerations. 

Additions:
Ensure the PRD reflects a user-centric perspective, considering the needs and experiences of the target personas. Use a clear and professional style that is easy for both business and technical teams to understand.

Type of Output:
A structured PRD document with clearly defined sections, formatted for ease of reading and reference.

Extras:
I want you to help me complete this PRD by asking me questions in an interview format where we can work on this back-and-forth and sequentially through each section. I might not have all the answers for you right now so it would be helpful if you also ask exploratory questions and help me fill out the section to the best of our abilities. If I cannot answer your exploratory questions, give me the option to skip the question so we can come back to the outstanding question later.  Incorporate placeholder comments for any unknowns or TBD sections. 

Using Generative AI in the workplace

Interestingly, there were a number of people asking for use cases for Generative AI in the workplace. I think this is still a case of “okay, this cool technology exists, it is game changing, now tell me how it can do all my work for me.”. That is really not an effective approach or mindset for driving adoption of generative AI.

Even GitHub Copilot in its current iteration is unable to do all the work for you. The work has to come from somewhere and it has to be validated by someone.

For the AI enthusiasts, I advocate that you experiment with it in the workplace. Pick an AI prompting framework, CREATE as an example, and try it out with your particular use case. Blog, demo, and make yourself available to help teammates make use of Generative AI. We all learn best by doing.

What is CREATE?

CREATE is a 6-part mnemonic for text generation prompting. The letters represent Character, Request, Examples, Additions, Types of Output, and Extras. Have a look at the prompt guide by Dave Birss for more information. He created one of the first LinkedIn learning courses back in 2022 which was my starting point to learning about AI text generation.