Building a Podcast Using NotebookLM
A Step-by-Step Guide On How to Leverage AI Tools to Create Engaging and Informative Podcasts
Introduction
Podcasts are rapidly becoming one of the most accessible and engaging ways to share ideas and insights, particularly in niche areas like procurement and automation. But creating a podcast that resonates with your audience can seem daunting—especially when juggling a busy schedule.
In this article, I’ll explain how I use NotebookLM, an innovative AI tool from Google, to streamline the podcast creation process. Whether you’re a procurement professional looking to share trends, discuss automation, or highlight key challenges, this guide will show you how to produce high-quality podcasts efficiently.
Step 1️⃣: Gathering Source Materials
The foundation of a successful podcast lies in the quality of your research. To start:
Collect Relevant Materials: Gather procurement reports, case studies, industry articles, and other materials related to your topic. If you don’t have materials, use tools like perplexity.ai to search for high-quality content.
Organize and Review: Next, you evaluate the relevance of the collected materials, filtering out anything that doesn’t align with your podcast goals.
Diverse Sources: Include a mix of PDFs, articles, YouTube videos, and website links to provide varied perspectives and enrich your podcast content.
Pro Tip: Use procurement-specific keywords like "indirect procurement efficiency," "automation trends," or "AI in procurement" when searching for materials” to ensure targeted results.
Step 2️⃣: Uploading to NotebookLM
Once you’ve curated your sources:
Upload the materials into NotebookLM.
One of the tool's fantastic features is that it supports multiple file types, allowing seamless integration of various content formats.
Organize materials into thematic Notebooks (e.g., "AI in Procurement" or "Procurement Trends 2025").
Why This Matters: NotebookLM uses advanced algorithms to analyze your content and extract key themes, helping you craft a podcast narrative efficiently.
Step 3️⃣: Instructing the Tool
Clear instructions are key to generating relevant content. When configuring NotebookLM, include:
Audience: Specify whether your audience includes procurement professionals, CFOs, or small business owners.
Themes: Focus on emerging technologies, cost-saving strategies, or efficiency improvements.
Tone and Style: Ensure the tool understands your conversational yet professional tone preference.
Example Prompt for NotebookLM that I use from my prompt library:
Things to Try: - create a highly engaging and well-structured 10-minute podcast on the [TOPIC]. Highlight no more than four relevant trends clearly and with relevant examples. - the audience is highly engaged and experienced procurement leaders.
Step 4️⃣: Evaluating the Podcast
Once NotebookLM generates the podcast, review it thoroughly:
Listen for Relevance: Ensure the content aligns with the intended audience and theme.
Check for Clarity: The script should flow naturally and avoid overly complex jargon.
Focus on Engagement: Does the podcast keep your interest? Are there actionable insights?
If you identify issues, move to the next step.
Step 5️⃣: Refining Input if Needed
Not all first drafts are perfect. If the output lacks focus:
Write a note to share your perspective on the materials or provide relevant examples you want to be included in the podcast.
Add this context to the Notes section of NotebookLM.
This iterative refinement ensures your podcast meets audience expectations.
Pro tip: I use a speech-to-text GPT I created to articulate my perspective on the materials and produce written material to copy and paste into the Notes section of NotebookLM.
Step 6️⃣: Revising and Improving
Improvement is an ongoing process. To create a polished podcast:
Edit for Clarity: Refine complex ideas into simple, digestible content.
Integrate Feedback: Use insights from your first review to enhance the next iteration.
Recreate the Podcast: Upload your revised materials into NotebookLM and generate a new version.
Case Example: For a podcast on "Procurement Trends 2025," the first version was 26 minutes and overly broad. By refining the source materials with additional context, the second version was 10 minutes and more concise, aligning better with audience needs.
Example Use Case: Procurement Trends 2025 Podcast
Step 1️⃣: Here is my query to the source material
Here is the link to my query.
Step 2️⃣: I uploaded the source materials
2 YouTube Videos
2 pdf files
1 website
Text from one website
Step 3️⃣: I instructed the NotebookLM
“Things to Try:
“- create a highly engaging and well-structured podcast on the on procurement trends. Share relevant examples.
”- the audience is highly engaged and experienced procurement leaders.”
Note that I did not specify the length or number of trends.
Step 4️⃣: I evaluated the podcast
Below is the result. At Minute 11, I felt like I had heard it all but realized I wasn’t halfway through the podcast. In short, it was too long.
Step 5️⃣: I refined the source materials
In NotebookLM, I used the Chat function to pull out the five most relevant trends. I then converted the note to the Source material with the click of a button. (That's a very nice feature: Kudos to whoever came up with that!)
Step 6️⃣: I revised my instructions to NotebookLM
Things to Try:
- create a highly engaging and well-structured 15-minute podcast on the on relevant procurement trends for 2025. Highlight no more than 5 key trends in a clear manner and with relevant examples.
- the audience is highly engaged and experienced procurement leaders.
Here, the result is the output. You be the judge — but I was a a lot happier about this!
Conclusion
Creating a podcast might seem challenging, but tools like NotebookLM make it achievable and efficient. Following this step-by-step guide, you can leverage AI to share insights, build thought leadership, and engage your audience.
New articles are published every Thursday! If you found this guide helpful, consider subscribing to "Zen and the Art of Procurement Automation" on Substack. Share your thoughts in the comments—your feedback could inspire the next podcast topic!