If you haven't heard of Notebook LM in 2025, you're wasting hours of your time every single day. It's a free AI tool by Google that can analyze websites, files, and even YouTube videos for you in seconds. You can then ask your questions, create summaries, and even audio overviews that you can listen to and interact with.
But it doesn't stop there; Google has just dropped Notebook LM Plus for business, which allows you to do crazy things, like creating no-code AI agents. I haven't seen companies use it to its full potential just yet, which is why I'm making this video. So let me show you how to use it so you can crush your competition this year.
If you've never used Notebook LM before, you'll first need to learn the basics. Let me quickly show you how the free version works before we get into using the Plus plan for your business. If you already know how to use it, feel free to skip to this timestamp over here or just stick around for useful tips.
A notebook is kind of a library for the topic you want to explore. It uses the information you upload as a knowledge base. So, for example, let's say you want to learn everything about growing on LinkedIn and you want to create a comprehensive strategy.
You'd create a notebook for this topic, add sources like YouTube videos from experts, and article links, and your notebook is ready. You can view your notebooks on the dashboard like this, or as a list view, and you can even filter them by title, creation time, or choose to see shared notebooks that will be available on the Plus plan that we'll see later in this video. If you prefer a different look, you can also switch between the dark and light themes in the settings.
So let's say you're a business owner deciding whether you want to buy Google Suite or Microsoft 365. For me, the answer is obvious, but maybe for some people, it's not. So instead of spending hours researching both platforms, you would just click "Create a Notebook.
" You would need to upload your sources, copy links from videos or articles, and paste everything here. It might take a few minutes, but considering you won't be reading or watching anything yourself, you're still saving a ton of time. If your sources don't load due to restrictions, you can just copy the text and paste it manually as pasted text.
For videos, it’s kind of the same thing; you can just transcribe them with an extension like this and paste the transcript. Just don't forget to rename the sources and delete bad links to stay organized. Currently, links have to be added manually one by one.
I wish there were bulk options, so fingers crossed Google will add it soon. Once you’ve uploaded your knowledge base, it’s ready, and now we can do whatever we want with it. We can ask it questions, for example, and it will provide us detailed answers with annotations leading to specific sources.
This ensures it’s pulling real information and not just AI-generated content. Once you open a source from annotations, you’ll see that it has a very neat summary from Gemini, and you can even watch videos directly from the source. Apart from directly asking AI questions, you can also create documents from presets, like for example, you can ask it to create a study guide, a brief, a timeline, or an FAQ—which is my favorite because it kind of generates common questions and answers, and you can learn so much from that.
If you like the content generated, you can save it as a source for the future. While you're learning and kind of going through the notebook, you can also add your own notes, which is a lovely option. You can also format them as you like.
And now, the feature that blew everyone away: audio overview! Once all of your sources are added, you can click "Generate," and in a few minutes, it will create a human-sounding podcast that is actually super interesting to listen to. In 15-20 minutes, you’ll learn everything from your knowledge base hands-free.
Let’s just hear a tiny piece of the podcast that was generated. One article suggested that Microsoft might be edging ahead in this AI race, highlighting how Copilot can practically build entire presentations just from a few prompts. Imagine feeding it a 50-page market research report and getting back a concise bullet-point summary, like in seconds!
That’s the power they’re talking about. Microsoft’s definitely making bold claims, right? It does show Copilot’s potential.
Isn’t it amazing? The podcast was so good, and it highlighted the most important things, like the use of AI, which is super important nowadays. But what's even cooler is that now you can interact with this podcast.
So you would click "Interactive Mode" and just start listening as you normally would. If you have a question, you just need to click "Join" and ask it. Now let’s see how it will answer.
My question is, “Is there really a big difference in pricing? Everybody says that Microsoft Office 365 is much cheaper, but I don't see that difference on the website. ” That’s a great question, and it’s easy to get confused by the pricing.
Yes, it can be a bit tricky to compare. What's fascinating is that at the entry level, both Microsoft and Google start at about $6 per user per month. Right?
So at first glance, they seem to be the same. But here's where it gets really interesting: Microsoft throws in one terabyte of OneDrve storage with that basic plan, which is a ton of storage. How amazing is that?
The answer is so relevant, and the information is just amazing! Now, if you want to listen to this podcast on your phone, you can just go straight to Notebook LM and log in, and you'll get access to it. Or, you can download it if you have a very old phone or something.
So now you know pretty much everything you need to know about Notebook LM, and we are ready to get into the fun stuff and build some AI agents. But let's first see what the Notebook LM Plus plan has in store for us. Well, firstly, you can customize the answers by giving your bot instructions on how to respond in a specific way.
So, instead of just giving us what it thinks is right, it will respond exactly the way we want. You also get shared notebooks, so you can collaborate with your team without re-uploading materials and links. Plus, we get analytics to see how the team is using this tool.
Listen to this: with a Plus plan, we get five times more notebooks, sources, and queries, which is pretty much 300 sources per notebook. So you can upload a ton of information, and we also get increased security, which basically means that our materials will never be used to train AI models. Therefore, we can upload important documents and confidential information if we need to.
There are, of course, a ton of things we can do with these features, like transcribing meetings, analyzing reports, and documents. The possibilities are really endless! But the coolest thing about it is that we can create custom AI agents with no code for pretty much any task.
It would be perfect for tasks like sales, onboarding, marketing, and even development. As it says on Google's website, the way we can create these agents is by adding specific notebooks and giving them a detailed knowledge base in the form of articles, documents, and videos that we already have. Then, we adjust the way they answer by giving them specific instructions.
So, let's create an AI agent with Notebook LM for onboarding, for example, and I'll show you exactly how it works. But before you start playing with it, you'll need to get the Notebook Plus plan. You can do it through Google Workspace or Google Cloud.
If your company already uses Gemini Business, you might already have access. Just log into Notebook LM, and if you see analytics on the top, you're already on the Plus plan. Just make sure you do it with your Google Workspace account and not your personal one.
If you don't have it, you will need to sign up for Gemini Business or Enterprise through your Workspace. So, for our example, I'll actually be creating an onboarding bot that will help our new employees learn how to use our calendar software really quickly. This bot will allow people to learn really quickly, ask questions anytime, and troubleshoot issues without bothering the IT department every time.
So step one in creating our agent is to actually create a notebook and upload a knowledge base. For this example, I'll just upload a ton of Google Calendar support pages. But of course, as a company, you would upload your own knowledge base and internal documents.
Uploading files might take some time, but remember, you only need to do this once, and later on, if you have some new information, you can always update it. Step two is setting our language instructions. Once all of the files are uploaded, we can now define how the bot responds.
By clicking on this toggle sign, we can define our conversational style as an analyst, guide, or by giving it custom instructions, which is what I'll do here. In the instructions, I'll tell the bot to respond as an IT specialist, keep answers concise and step by step, and also to provide links to relevant sources whenever possible. I'll also select the shorter response length, so the answers are pretty straight to the point.
And just like that, a bot is created, and we don't need to do anything else apart from sharing the link with our team. Just click the share button at the top of your notebook and add all of the emails. Note that you can only share notebooks with the members of your Workspace organization, so it won't work with random emails; those have to be your team members.
So, let's say I have a new employee who needs to learn about this software. I'll share this bot link with her, and she can now access the bot from anywhere. She can generate a podcast and listen to a summary in just a few minutes, and she can also ask questions while listening and get instant answers.
Let's hear a quick example: "How can I require payment for calendar events? " To require payment, you have to connect a Stripe account to Google Calendar, which you can do right in the settings. Then, when you create your appointment schedule, you set the price, and customers pay when they book a time slot on your page.
So it's all handled by Stripe, and you can even add a cancellation policy to manage expectations. So let's say our employee is more settled now and works with this software every day, and now she just runs into an issue. Instead of waiting around for it, she can simply ask the bot to provide a quick, straightforward answer, and it will give it to her on any issue.
And just like that, instead of hiring ten IT guys, you just need one to manage all of the AI agents. So, learn AI, guys. But all jokes aside, you can really create as many of these agents as you like and add a ton of information to them to really streamline your workflow without any code for $20 a month.
If you'd like me to make. . .
A specific video, please! Make sure to leave your suggestions in the comments; I'd love to know what you guys want to learn about. If you've enjoyed this video, please give it a like and subscribe.
It will really help this channel, and I'll see you in the next video.