Before “just ChatGPT” became popular, Google wasn’t the first place I turned to for answers. It was YouTube. Seeing someone demonstrate how to do something just clicks in a way that a wall of text never could. This may seem ironic coming from a writer whose job is to write things that people are supposed to read rather than skip to a video, but I know I speak for the majority when I admit that short videos win most of the time.
However, the important term here is quick. Unfortunately, YouTube videos are much less quick. And that’s exactly where NotebookLM comes in handy. Turn hours of video into a knowledge base that can be queried in seconds.
There’s too much filler on YouTube these days
Since when did all videos on YouTube get this long?
Just as LinkedIn, Reddit, and unfortunately many articles (I promise this isn’t one of them) have become mere AI-generated stitching, YouTube is not immune. Finding videos that actually match the query you have in mind is always hit or miss. You won’t know if the video actually answers your question until you watch it for a few minutes. If you’re lucky and find the answer in the first video you click, that’s fine, but that rarely happens.
There are so many AI-generated slops on YouTube now that the chances of your first video being the correct one are lower than ever. Sometimes, one helpful video may not contain all the information you’re looking for. It might answer one part, but the rest is split into several other videos that you’ll have to find and piece together yourself. That’s exactly the kind of busy work that NotebookLM can help you with.
Before I get into the details of the workflow, I would like to mention that this is only useful if you watch YouTube for the purpose of learning something, not for leisure. This is by no means a replacement for traditional YouTube. No one runs comfortable videos or great video essays through NotebookLM. This is strictly to filter out the tedious tasks, the tutorials and explanations you watch to extract information, and the videos you should actually spend your time on.
NotebookLM reads the video, so you don’t need to read the video.
That wasn’t the case as I made it all the way through the tutorial
For those who haven’t used it, NotebookLM is a source-based AI tool, and this is the key detail that makes this whole thing work. We answer only from the material given to us, so we don’t guess, hallucinate, or pull random nonsense from the open web or internal training data. One source type that NotebookLM happily accepts is YouTube videos. When you drop in a link, NotebookLM takes the transcript of the video and treats it as text that you can search, interrogate, and cross-reference.
That’s all the trick. Add a single video or an entire pile, and query large amounts of videos with friendly language prompts. Ask what you really want to know and NotebookLM will answer you directly from the video transcript. This is a great way to consolidate information. This is especially effective if you have multiple videos covering the same topic from slightly different angles, and if you combine those videos with other sources within the same notebook.
It’s also a great way to sort out which videos are worth your time and which aren’t. When you start asking Notebook a question through the chat panel, you’ll quickly see which videos have the answers you need, and which 10-minute videos you can skip entirely. All answers returned are backed by clickable quotes, so you can see anything right away. Quotes are linked directly to the relevant passage in the video transcript, so you don’t just take the AI’s word for it.
I no longer “watch” YouTube videos. I’m using NotebookLM to watch.
NotebookLM changed the way YouTube works.
NotebookLM, so you can incorporate the rest of your toolkit into the mix. If you want to digest everything in podcast format, you can turn a stack of videos into an audio summary or generate a mind map to see how topics connect between creators. All answers returned are backed by clickable quotes, so you can see anything right away. If you want to take it a step further, you can generate an overview video or a slide deck summarizing your key points. The options are endless!
The best part about this workflow is how seamless it is.
Simply copy the video link and drop it into an existing notebook or a new notebook. that’s it. And if you’re working with lots of videos at once, there are plenty of Chrome extensions that allow you to bulk add YouTube videos to NotebookLM, so you don’t have to paste links one by one.
