Image by editor
# Introduction
creating a Product Requirements Document (PRD) is a common process product management And a common work in fields such as software development and the tech industry as a whole. Some of the difficulties and stringent requirements commonly found in creating a PRD include ensuring clarity, preventing scope creep, and preserving stakeholder alignment.
Thankfully, AI tools have emerged to help tackle these challenges more effectively, without completely handing over the strategic decision making underlying the PRD creation process – in other words, with the human still in the loop. is an example Google’s NotebookLMWhich synthesizes grounded raw data or materials to answer questions, thereby turbocharging the workflow to create grounded, useful PRDs.
Based on a beginner-friendly use case, this article will guide you through the process of using NotebookLM’s features to transform raw, sometimes chaotic information into grounded PRDs in minutes. spoiler:It won’t just be about chatting with an AI assistant.
# From messy notes to structured PRD drafts
Let’s consider the following scenario. You are a newly hired product manager for a startup that wants to develop a new mobile app. florafriend. The goal of the app is to help people prevent accidentally killing their houseplants.
The team, including you, has put together a set of three “untidy” documents that contain a description of what the potential app should be like:
interview_transcript_matt.txt: A 30-minute interview with a user named Matt, who owns over 50 plants. In these interview notes, Matt says that existing apps are “overly complex” and make it difficult to keep track of aspects like “which fertilizer to use.”competitor_research_notes.txt: A thick list of bullet points created after analyzing competing apps like “PictureThis” and “Planta”, highlighting their paywalls and interface shortcomings.brainstorming_whiteboard.jpg: Random but somewhat “good” ideas that are mentioned by the team during lunch breaks and other casual conversations, for example “make a Spotify playlist for plants”, “set watering reminders”, etc.
Imagine complete documents containing all the content described above. Converting these manually into a clean PRD that brings everything together nicely can seem like a pain, right? Enter NotebookLM!
enter into notebooklm With your Google account and click “create new notebook“. Give your new notebook a name, something like “FloraFriend PRD“
Once the new notebook is created, you will be welcomed to the main NotebookLM interface, which looks like this:
notebooklm interface
A word of caution: This newly minted notebook is not intelligent in itself. it’s not regular big language model (LLM); It does not know plant care or any other specific topic. But we’re going to teach an “express” master’s degree about it with our messy – yet enlightening – notes for the tool.
Let’s say you have the above three files containing some content related to a Plant Care app, or any other raw information files of your own. You can upload them to the NotebookLM canvas using the Upload button in the main, central section.
Once uploaded, you can treat your notebook to a small, toy-sized recovery-enhanced generation (RAG) systems that can start thinking and behaving like AI based on the information they have access to. In fact, without asking, by clicking on one of the uploaded files on the left, NotebookLM produces a short, well-organized summary of that file’s contents: It’s called File source guide.
Now comes the main part. We can just ask something like “Write a PRD” in the chat box below, and that’s it. But we want to do it right and provide clear, specific instructions, and that involves some quick engineering, namely forcing the nascent AI to prioritize what we want our PRD to reflect: prioritizing user problems over random ideas generated by the team (without ignoring them completely). Here’s a well-crafted prompt that works:
I’m a product manager for FloraFriend. Prepare the draft PRD based on these sources only.
Important Constraints:
1. Prioritize features that solve the problem points outlined in interview_transcript_mat.txt.
2. Eliminate any ‘brainstorm’ ideas that don’t directly solve the user’s problem.
3. Structure the output with these headers: problem statement, key features, non-functional requirements (UI/UX), and success metrics.
Try adapting this hint to your business problem or use case. Once sent, chances are you will get a nice and clean PRD with key sections like problem description, key features, non-functional (UI/UX) requirements, success metrics, etc.
Interestingly, the PRD has something that looks like numerical quotes that you can hover over. If you do this, you’ll see the source (one of the source files) pop up:
Before accepting this first PRD as is, remember that the first draft is rarely perfect. Keep engaging in the conversation to gradually refine it, for example if you notice a monetization section is missing, ask: “Based on competitor_research_notes.txt, what monetization models are our competitors using, and what should we avoid?“. After that, check the outputs manually, make sure they are consistent with the rest of the first PRD draft, and include key monetization insights in it, either manually or ask NotebookLM’s AI to do it – if you opt for the latter, always check what you get before blindly approving it. Remember: AI can make mistakes!
the icing on the cake is this audio overview Section (Studio) on the right-hand panel. By simply clicking on it, you will produce an audio overview of the information contained in the source files. This is an excellent way to assimilate information when reading may be less engaging, for example when you are on a daily commute.
# next steps
This article introduces NotebookLM’s capabilities to generate grounded PRD specifications from raw, dirty documents in just a few minutes in very simple steps. From here, a meaningful next step can be taken Google’s antigravity To transform your PRD specification into a functional software prototype.
ivan palomares carrascosa Is a leader, author, speaker and consultant in AI, Machine Learning, Deep Learning and LLM. He trains and guides others in using AI in the real world.
