YouTube + NotebookLM: The Marketing Magic Formula ✨

EDITOR’S NOTE

Dear Future-Proof Humans,

What if your AI assistant could...

1️⃣ Analyze the sentiment across thousands of YouTube comments instantly?

2️⃣ Identify key consumer feedback on specific products or topics?

3️⃣ Spot emerging trends and generate ideas directly from audience discussions?

We've been exploring AI workflows for a few weeks now.

After showing you how Claude can work with email management, design automation in Figma, and various MCP connections, I wanted to design a workflow that analyzes unstructured data, such as hundreds of YouTube comments, to generate actionable insights.

In this edition of Nanobits, we will do:

⚡ A walkthrough of setting up the YouTube comment scraping workflow with NotebookLM

🔧 See some sample prompts you can copy for your own market research

⚠️ Explore practical ways to use this analysis for competitive intelligence on any product or market

So, let's turn those thousands of YouTube comments into actionable insights.

Let's begin!

CONSUMER SENTIMENT ANALYSIS FROM YOUTUBE COMMENTS USING APPSCRIPT & NOTEBOOKLM

The core of this workflow lies in getting the vast number of YouTube comments into a format that NotebookLM can easily digest. The key is using Google Docs and Google App Scripts.

Pre-requisites & set-up

Here's what you need before getting started:

• NotebookLM Access: You need access to the NotebookLM tool.

• Google Account: A standard Google account is required to create Google Docs and use Google App Scripts.

• Google App Script Code: You need the specific code for scraping YouTube comments into Google Docs. This is often provided via a shared document or resource linked by the workflow creator.

• YouTube Data API V3 Enabled: You need to make sure the YouTube Data API V3 is enabled in your Google Cloud project associated with the App Script.

• YouTube Video IDs: The IDs of the specific YouTube videos whose comments you want to analyze.

Here's the breakdown of how to set up the YouTube Comment Analysis Workflow and configure it to work with NotebookLM:

Step 1: Create a New Google Doc

Start by creating a new, empty Google Doc. This document will serve as the container for the YouTube comments you scrape. Give it a descriptive name, like "GaN Charger Review Video Comments - [Video Title]".

Step 2: Access App Script

In your new Google Doc, go to Extensions > App script. This will open a new tab where you can write or paste custom code. Think of App Scripts like writing code (similar to VBA in Microsoft Office) that runs inside your Google applications and can interact with other services, like the YouTube API.

Step 3: Replace the Default Code with the Comment Scraper Script

You'll see a default code area. This is where you paste the script that will download the YouTube comments. The script was generated with the help of tools like ChatGPT. While you don't need to be a coder to use it, having a basic understanding can be helpful if you ever need to adjust it.

Step 4: Enable the YouTube Data API V3

The script interacts with YouTube's data using the YouTube Data API V3. This API must be enabled for the script to function correctly. In the App Script editor, you'll need to make sure this API is enabled in your project settings.

Step 5: Save, Authorize, and Run

Save your script project and go back to your Google Doc. Refresh the page, and you should see a new custom menu (e.g., "YouTube tools"). The first time you use it, you'll need to authorize the script. Once authorized, clicking the menu option will prompt you to enter the YouTube video ID.

The script will start downloading comments directly into your Google Doc. A few hundred comments can be downloaded in seconds, while thousands might take a minute or more. Once finished, your Google Doc will contain all the scraped comments with a summary at the top and direct links to each comment.

Step 6: Add the Google Doc to NotebookLM

Now that your Google Doc is populated with comments, it's ready to be analyzed. Open your NotebookLM notebook and add this Google Doc as a new source. You should also add the transcript of the corresponding YouTube video, which NotebookLM supports natively. For competitive research on GaN chargers, repeat steps 1-6 for several relevant videos and add all the resulting comment docs and transcripts to the same NotebookLM notebook.

As you can see, NotebookLM now has full access to the video transcripts as well as all the comments.

TASKS WITH NOTEBOOKLM: ANALYZING GAN CHARGER REVIEW COMMENTS

With your sources loaded into NotebookLM (including video transcripts and the Google Docs filled with comments from relevant GaN charger reviews and comparison videos), you are ready to perform analysis using simple text prompts.

Let's apply this workflow to your competitive research scenario, focusing on GaN chargers in the Indian market.

Task 1: Understand Overall Consumer Sentiment

Starting with the overall sentiment is straightforward. You want to know how consumers generally feel about GaN chargers, specific brands (e.g., Anker, UGREEN, Croma), or particular models mentioned in the comments of videos targeting the Indian audience.

Prompt: 

Analyze the video scripts and comments thoroughly and summarize the overall audience sentiment regarding CMF 65W GAN Charger by Nothing and other competitor brands. Break down the sentiment into positive, neutral, and negative tones, and explain what this indicates about consumer perception with 3 examples for each sentiment.

What NotebookLM Returns

NotebookLM will analyze sentiment across your comments and provide percentage breakdowns (e.g., 6% positive, 92% neutral, 2% negative) with supporting examples. The embedded links let you click directly to the original YouTube comments for verification and context.

Task 2: Brand and Product Analysis

When researching your product, it's important to understand what's working and what isn't from the consumer's perspective. This task helps identify actionable insights for brand positioning and product development.

Prompt: 

What are the top 3 things CMF 65W GAN Charger by Nothing should continue doing as a brand and Top 3 things CMF 65W GAN Charger by Nothing should stop doing as a brand. Support with concrete comments as quotes.

What NotebookLM Returns

NotebookLM will identify common praise points and complaints, categorizing feedback into "continue doing" and "stop doing" actions for brands. It provides supporting comment quotes with clickable links back to the original YouTube sources for verification.

Task 3: Discover Emerging Trends and Feature Ideas

Consumer comments are not just about current products; they often contain suggestions, questions, and discussions about desired features, alternative uses, or gaps in the market.

Prompt: 

Analyze the textual content of the video scripts along with their corresponding viewer comments, to identify emerging trends in GAN charger features and consumer expectations. Summarize the key features mentioned (charging speed, number of ports, heat management, etc.) and explain how these trends might shape the GAN charger market.

What NotebookLM Returns

NotebookLM identifies user feature requests, compatibility concerns, and unmet needs mentioned in comments, revealing potential product opportunities. It provides clickable links to original comments, so you can find specific user suggestions that might spark your next product idea (higher wattage multi-port chargers).

Task 4: Competitive Comparison Insights

If you've added comments and transcripts from videos comparing different GaN charger brands or models available in India, you can prompt NotebookLM for direct comparisons based on consumer experiences.

Prompt: 

What are the key differentiators between CMF 65W GAN Charger by Nothing and its competitor as expressed by viewers?

What NotebookLM Returns

NotebookLM compares how consumers discuss different brands, highlighting specific trade-offs such as Brand A's reliability vs. price or Brand B's value vs. heat issues. You get direct comparative analysis with supporting quotes and links to original comments.

TASKS WITH NOTEBOOKLM: PULSE CHECK ON THE RECENT GOOGLE I/O 2025 LAUNCHES

I recently watched The Verge’s interview with Google’s CEO, Sundar Pichai, on the future of search, AI agents, and selling Chrome.

I quickly conducted two analyses on the video: one to determine how these trends (AI agents, search evolution, Chrome browser, web browsing changes, etc.) might shape the search and AI market, and the second to gauge general consumer sentiment about the recent AI innovations launched by Google.

For instance, if I were the brand or product marketer at Google, the following insight would be instrumental in redefining the messaging and positioning framework for certain Google products.

In the next prompt, I asked:

Analyze the comments thoroughly and tell me 3 things that people generally liked about the Sundar Pichai interview, what more would they have wanted to hear discussed, and any new business ideas or opportunities mentioned by viewers.

These are some fantastic ideas suggested by users in the comments section, which can be analyzed by Google's product teams to identify which ones to add to their roadmap.

IMPORTANT LIMITATIONS & TIPS TO BE AWARE OF

While this workflow is useful and accessible, keep these limitations in mind:

• Google App Script Execution Time Limit: Scripts timeout after a few minutes, so videos with 10,000+ comments might only partially download.

• NotebookLM Retrieval Limitation: For very large documents, the free version of NotebookLM processes selective parts rather than every single comment, which can affect accuracy for complete document analysis.

• YouTube Data API Quota Limits: You get 10,000 units per day (roughly 1 million comments), though typical usage rarely hits this limit.

• Automatic Syncing with Google Docs: NotebookLM automatically updates when you refresh the source, so you can re-run scripts to add new comments and keep analysis current.

End Note

That's a wrap on this deep dive!

We've seen how you can connect raw YouTube comments to NotebookLM, transforming a manual task into an efficient AI workflow. This approach turns simple prompts into detailed consumer insights with proper references back to the original sources.

Try connecting Google Docs filled with comments from your target videos to NotebookLM and see what happens when AI reads and synthesizes consumer opinions for you.

Three YouTube Comment Analysis experiments to try this weekend:

• Scrape comments from a review video of a competitor's product and ask NotebookLM to "Summarize the main criticisms and praises mentioned by users."

• Download comments from several comparison videos covering products in your market and ask NotebookLM to "Identify common questions or confusions users have when comparing these products."

• Analyze comments on videos discussing a broader trend related to your market and ask NotebookLM to "Identify any mentions of desired features or frustrations related to your product/solution in these discussions."

Start small. If you get stuck, reach out! Thanks to the YouTube creator, The AI News, for the comment scrapper script.

Which competitive analysis or market research tasks would you automate first using this workflow? Let me know what consumer insight challenges you'd like AI to solve for you.