Business Guide to NotebookLM

Everything you need to get up to speed on NotebookLM and get you started with using this powerful AI by Google.

Hey — It’s Hussein 👋

Today, I’m really excited to finally write about NotebookLM. I’ve been wanting to write about this for a while, and I’m happy I was finally able to free up the time to do it.

NotebookLM is like a super-powered assistant that can take a mountain of data and give you summaries and briefs in multiple formats, plus a mind-blowing, podcast-quality discussion that summarizes everything you upload.

Listen to this podcast, which I created with a click of a button, no-prompt required to see what I mean! 🤯

Before we begin, a quick note on my latest side project, Startup Beach:

  • Create a profile and get matched weekly with a community member for an in-person or virtual meeting. It’s a great way to grow your network.

  • Host your next event on Startup Beach and take advantage of our matchmaking and community features.

What is NotebookLM?

NotebookLM is an AI tool by Google that lets you upload and link a bunch of information (sources) and then get structured info based on them like summaries, timelines, briefs and more. Plus it gives you the ability to ask questions and create a podcast-like audio brief!

When you first open NotebookLM, you’re greeted with a few sample notebooks, a barebone UI that feels like another LLM-type tool, and not much else. Don’t let that stop you from really giving NotebookLM a shot because it is mindblowing.

Things get really interesting when you start you own Notebook and add your own sources. So get to that step ASAP!

Think of any project or product you are currently working on and add a bunch of links, PDFs, docs whatever you have just to give it a try. Then tap on the text: Notebook guide to open the menu of capabilities.

Where all of the magic happens

For your info, and without getting too technical, NotebookLM uses a technique called RAG (Retrieval-Augmented Generation) to create its summaries. This means it retrieves relevant info from the sources you upload before generating a response. Think of it as a mini training ground for the LLM based on the information most relevant to you plus the general training data of the model itself.

How does it work?

First, you need to feed it information. You can do this by uploading:

  1. PDFs, text files, audio files

  2. Google Docs

  3. Google Slides

  4. Links to online articles

  5. YouTube videos (it transcribes them)

  6. Text you paste or type directly

Right out of the box, here’s what you can do with that info:

  1. Generate an auto FAQ.

  2. Create a study guide.

  3. Build a table of contents.

  4. Create a timeline of events.

  5. Draft a briefing document.

  6. Ask questions.

  7. And last but not least, create an audio overview in podcast format, where two AI voices discuss the content, complete with jokes and questions.

What can you use it for?

As they say the possibilities are endless but that doesn’t help you much. So here are 10 business use cases for NotebookLM to get your ideas flowing:

1. Prep for meetings:
Imagine you are about to go into an important meeting with a high-profile individual, preparing for an interview, or pitching a big client. Instead of manually reading endless articles on the web, you can upload the articles or links to them, YouTube videos, and more. Then, you can quickly create a timeline, a briefing document, and even a podcast to listen to while driving or flying over to the meeting - if you are back to in-person meetings, that is.

2. Prep for podcast interviews:
If you are a podcast host and need to prep before interviewing someone, you can upload all your research materials and get a briefing and list of questions before your interview.

3. Absorb knowledge from books:
If you have PDF versions of books, you can upload them, get summaries, ask questions, and digest those books in a much quicker and more efficient way. We are getting close to that scene from The Matrix where you can just plug a cable into your head and download new capabilities and knowledge.

4. Chat with your PDF or any document:
Quickly ask questions about contracts or other agreements that you upload. (Reminder: always get a final check with your lawyer!)

5. Create summaries of documents or topics:
Upload all your documents and ask NotebookLM to create a summary. This is great for summarizing research or academic papers.

6. Onboard new team members:
Upload all your PRDs, requirements, design docs, and technical write-ups. NotebookLM will generate summaries for new members to read and allow them to ask questions about the docs. This is a great way to get new team members up to speed very quickly.

7. Create podcast-like audio files:
While I wouldn’t necessarily publish these (for now) if you are a better listener than reader or the timing works better to listen than read, then you can upload content and have it create a custom podcast just for you. A clever hack is to upload links to texts or videos on topics you follow each week and get a short, 10-15 minute podcast that catches you up on what’s new over the week in podcast format.

8. Replace your note-taking app:
You could replace apps like Notes, Google Docs, or Evernote with NotebookLM to take advantage of the AI features. I know habits are hard to break. I am still using Apple Notes, but I’m considering giving this a try. If you try it, let me know how it goes!

9. Use it as a brainstorming partner:
Upload information on a topic and let NotebookLM suggest ideas, perspectives, and fresh discussion points.

10. Create SOPs for your teams:
It’s great for creating Standard Operating Procedures for your operations, DevOps, or member success teams.

Quick note on data privacy

Just in case you are concerned. Google has made it very clear that the sources and information you upload are not used to train their AI model. I would still advise you to not add any personally identifiable information (PII) tho and remember to always follow your company’s AI usage guidelines when it comes to data sharing concerns.

What are the limitations?

We can’t ding Google too much for these since it’s still an experiment tool that is under construction but here are some limitations to keep in mind:

  1. It’s missing support for critical file types like CSVs and Sheets/Excel, which would be useful for deep analysis.

  2. Like all large language models, it can still generate inaccurate or made-up information (i.e., Hallucinations), though it should be less likely since it’s limited to the sources you upload.

  3. It doesn’t have a live sync with Google Docs/Slides yet.
If you make changes to a document, you’ll need to re-upload it into NotebookLM.

It’s wild to think that even with all of the features and use cases I described above, NotebookLM is still considered an experimental feature from Google and, as far as I know, is not heavily promoted yet.

I can’t wait to see what comes next.. will it have acess to my Calendar, Contacts and Gmail any time soon? Imagine the power then!

Let me know if you come up with any new use case for NotebookLM. I would love to hear about them.

Until next time!
Hussein ✌️

P.S. If you enjoyed this email, please forward it to a friend. (Use your custom link 🎁).

P.P.S. If you live in or visit San Diego, sign up here to be notified of startup mixers I organize in the area. Would love to meet you there.

How did you like today's newsletter?

Login or Subscribe to participate in polls.