# Did Claude Just Kill NotebookLM With This HIDDEN Feature?

## Метаданные

- **Канал:** Corey McClain
- **YouTube:** https://www.youtube.com/watch?v=CjEuLlfcnoc
- **Дата:** 28.02.2026
- **Длительность:** 9:32
- **Просмотры:** 9,120
- **Источник:** https://ekstraktznaniy.ru/video/10567

## Описание

START HERE TO WORK WITH ME. Create client-attracting content in one focused hour: clarify your message, generate a high-retention content, and repurpose it into ready-to-post assets → https://www.incomecreator.pro/ltao1?utm_source=youtube&utm_medium=video&utm_campaign=content-strategy-kit&utm_content=did-claude-kill-notebooklm

Claude Projects vs NotebookLM: The Research-to-Build Workflow That Creates Business Assets

In this video I show how Claude Projects differ from NotebookLM by going beyond storing sources and answering questions to actually building deployable business tools from uploaded files. The creator describes uploading hundreds of files into a Claude Project, passing a detailed implementation plan, and having Claude build a funnel-math and conversion-rate calculator that identifies revenue leaks and provides fixes, plus profit-first reports and snapshots. It outlines how Claude Projects switch into RAG mode after about 5% capacity, similar to NotebookLM’s retrieval-augme

## Транскрипт

### Introduction []

If you've been using Notebook LM to research and then manually figuring out what to build, you're going to want to watch this entire video about claw projects because claw projects do not just store information and allow you to ask questions like people do with Notebook LM. I was able to upload hundreds of files into a single project. And from those files, Claw was able to build a tool for my business that told me exactly what I was overlooking, what to do to fix it. And that's something that nobody LM just can't do. If you want to learn more about the tools you see in this video, they're are part of my 1hour content strategy kit, a kit that I designed for professional creators to help you create content that attracts clients, the link will be in the description. Most people treat Claw projects just like Notebook LM. They upload some PDFs, they ask a few questions, and then they're done. And I guess that's okay if you're using Notebook LM because it does have a dedicated chat space. But there's so much more going on inside of Cloud Projects. The biggest problem with how people use projects is that they treat them like a Google Drive folder, they drop information inside of it, they ask a few questions, and they think that's what they're supposed to be used for. And anytime I mention these other use cases, people always tell me that's not what projects are used for, but it works so well. And if anyone tells you that a certain way for using AI is not the proper way, then you already know you shouldn't be listening to them because there is no right way. The only right way is the way that helps you get the result that you're looking for. And so I'm not saying that the way you're using projects is wrong. I'm just offering you some alternative ways to use it that may get you results that you like. Truth is

### RAG Mode Explained [1:33]

Claw projects are a great place for doing research. I took my notebook LM research wizard that I used to configure notebooks inside of notebook LM to give me more control over how notebook LM operates as well as the research discipline that I'm using around different topics. And I converted it so that it works with claw projects. And the reason I did this is because when you upload files to claw projects and you get beyond your 5% capacity, claw projects changes from contextual awareness and it enters into rag mode. Now what you need to understand about rag mode is that this is the same thing as notebook LM. It stands for retrieval augmented generation or in layman's terms it just means that the AI is sitting in a closed room. it's in a library instead of running around the whole world trying to find the answer. And so it's only going to focus on the context of the sources that you've provided. This means that once you go past that 5% capacity, claw starts acting like notebook LM in the way that it thinks and reasons. It's going to stick to the script. files. Claw projects does the same thing minus the citations, but it also reasons across the contents of your files and it allows you to execute in other ways that notebook LM just doesn't. Remember when Google first connected notebook LM to the Gemini chat and you were able to add any notebook as a data source and it was almost like these superpowers we got from this combination, this new workflow. Claw projects have had that same moment, but no one's made the correlation between the two and it's definitely there. Once

### Research Or Build [3:09]

your sources are in the project, you have two paths you can go down. You can research or you can build. And to be honest, most people only think about researching. And so here's what building

### Building Funnel Tool [3:21]

looks like. Building doesn't look like prompting. Building looks like delegation, passing off your work to the AI to actually do it for you while you go and do some other things. This tool looks at your funnel math. It looks at your conversion rates. And it tells you exactly where your money is leaking. And it also tells you how to fix it. And I'll be honest with you, it was really easy to build this tool because all I had to do was add the project where all of the necessary information for building this tool was stored. And then I gave Claude a detailed implementation plan for what I wanted and pressed enter. I went grocery shopping, came back an hour later, and this tool as well as others were done. So, let's say you're on YouTube, Instagram, Facebook, Tik Tok, LinkedIn, Pinterest, X. You got a podcast and a newsletter. It's going to take all of the views from all of your platforms and calculate average conversion rates from the traffic from your content to your website. Then, it's going to calculate how many site visitors you actually have. You can choose whether you have a single product or a money model. Then, it gives you two profit first reports. One of them is a 90-day report. The other one is a real-time window. so that you know exactly what you should be doing with your income so that you can actually grow your business and then you have a snapshot print button here. Claw projects didn't just do research for me. It actually built a tool around my business around constraints I was dealing with and around problems that I

### Knowledge Versus Assets [4:47]

need to solve. So the difference between Notebook LM and Claude is knowledge versus assets that you deploy, activity, things that you actually do. And for me, that's really big because information is everywhere nowadays. But implementation is the scarcity because a lot of people are learning with AI, but they're not doing anything with it. And so any tool that moves me closer to action is going to be preferred in my workflow. But the

### CEO Logic Workflow [5:13]

calculator is only half of it. Because the calculator is a tool that lives in the library. It's not the logic. The logic is my CEO. Because as a solopreneur, a lot of times you're working in your business and you never have time to work on it. You're trying to figure things out. And so it's like, what if you had someone who knew what you needed to fix and could tell you what to do so that you could just do it and then work on the business instead of working in the business all the time.

### NotebookLM Data Pipeline [5:40]

This is where Notebook LM really shines because I was able to go to YouTube and other platforms and gather all kinds of free publicly available data about growing your business, running a business, making these type of decisions. And then because of this Chrome extension I have, I could just click this download button. I could select all, go plain text or markdown, never PDF because PDFs cost a lot of tokens to read. Never use PDF. And then I can download them. And now I have all of these data sources to reuse as many times as I like. I upload them to the project and activate rag mode and I may use one of my research skills if I need to gather some special data before I start building to make some decisions about how I want to build. But then I come to the claw desktop app and I add a project in that has all of the data that I'm looking to build with. And so when my CEO reads that data that I hand to it, it's making decisions about my business based on world-class expertise knowledge. It's not guessing. It's not hallucinating. I can look at the numbers. I can do the math. I can check the math and I can instantly tell if this is hallucination or if this is actually tracking. So the same project

### Self Questions Breakthrough [6:50]

same sources, but a different instruction. Instead of doing research, instead of asking questions, I asked it to Bill. And so after I created the CEO logic, I uploaded the calculator as a tool. I uploaded my founder blueprint, my audience blueprint, and all of the other relevant information and data that I could. And then I started asking questions. But one of the things that really unlocked a new level for me is when I told the CEO to ask its own five questions that I wasn't asking and to also answer the same five questions. And that's when I got the breakthrough that I was looking for because it found a critical mistake that I was making and overlooking. That would make all the difference between me hitting my goals in the next 90 days or not. It didn't just tell me what was broken. It gave me an actual game plan to fix it. So, I know exactly what to do. I can track whether it's working or not, and I know what to fix. I'm not spending all day trying to figure that stuff out. And so

### Run This Workflow [7:46]

if you want to execute this same workflow, it's very simple. You start a notebook LM. You gather your data. You export all of that data. You upload that data to a cloud project. And then you're at a fork. You can either research or you can build. So sometimes you're going to want to research because you're in rag mode. But for me, I'm inclined to build. So I'm going to go to co-work on desktop and then I'm going to connect that particular project that has all of the data there. And then I'm going to paste in an implementation plan and let Claude Co-work my assets for me. and anyone watching this video can run that same sequence today. Now, it might not come out like mine because of my personal logic and library skill, but nonetheless, it will still be a good asset. And if you're wondering whether

### NotebookLM Versus Claude [8:29]

this actually replaces Notebook LM, let me show you what I think about that. Notebook LM creates amazing 2D content, infographics, slides, data tables, and it creates great 3D content, interactive podcast, those are all great. But with Claw Co-work and Claw projects, you're able to create 4D content, things that move through time, assets that streamline your [clears throat] business, assets that actually move the needle in life if you use them appropriately to be more productive. So, no, notebook element isn't dead. But for operators who want to build, who don't already know about claw projects and co-work, they're definitely going to want to start using this workflow. And

### Builder Mindset [9:07]

so at the beginning of this video, maybe you're a researcher and you thought that Notebook LM and projects were all the same and you use Claw projects just to do research. But now you're a builder and that's the shift that you need to make. You can research or you can build. But if you want the assets you build with claw projects and co-work to come out like mine and you want to get a copy of my logic and library skill, then make sure you check out this next