As with a lot of things, note-taking is something that has changed a lot with AI now in the picture. Now, I'm a full-time student who has obviously been note-taking long before AI was a thing, and at this point, I've tried pretty much every AI note-taking tool out there.

Frankly, most of them are flat-out disappointing. They're just regular notes apps with a chatbot bolted on, and nothing about the actual workflow of taking notes changes. NotebookLM is the rare exception, and the biggest reason it works so well is because it doesn't really try to be a notes app at all. I've been using it for all of my studying since it was a Google Labs experiment, and here are 3 workflows that are genuinely better than actual notes.

Audio Overviews as a "review pass" for studying

My notes finally talk back, and they have a lot to say

If you haven't used NotebookLM much yourself before but have heard of it in passing, I'm willing to bet the feature you were introduced to was Audio Overviews. It's admittedly the most "flashy" feature of the tool, and is what made it the viral hit it is today. NotebookLM basically lets you create grounded workspaces called notebooks where you can add sources (PDFs, docs, webpages, YouTube videos, etc.) and interact with them via the Chat panel or turn them into Studio outputs.

One of these Studio outputs is Audio Overviews, which are basically engaging and interactive podcast-style conversations between two AI hosts that break down whatever you've uploaded. You can pick from a few formats (Deep Dive, Brief, Critique, Debate), customize the focus with a quick prompt, and even hop into Interactive Mode to ask the hosts questions mid-episode like you're calling into a live radio show.

Now, Audio Overviews went viral for a good reason. The entire reason why it beats traditional notes is because it solves the when problem. With apps like Apple Notes or Notion, you can only really engage with what you've written down when you're sitting at a desk, actively reading through your own scribbles. That's a ton of friction, and it's honestly why most of the notes I take end up dying in some folder I never open again. Audio Overviews flips that completely. Instead of being chained to a desk, I can engage with my notes during the dead hours of my day like when I'm commuting to class, walking to the grocery store, working out, or even doing the dishes.

One workflow I do every Friday is to dump the week's lecture notes (I have separate notebooks for each course I'm taking), slides, and any PDFs my professor shared. I then generate an Audio Overview with a custom prompt instructing it to give me a focused recap of that week's material, highlight the concepts I'd most likely be tested on, and spend extra time breaking down anything that ties back to topics we covered earlier in the semester! I listen to this podcast whenever I have free time over the weekend, and by the time I sit down to actually study during the week, I'm already half-primed on the material! The parts I struggled to follow during the podcast are the exact ones I know to focus my active study time on. No notes app, AI-powered or not, can do that.

Mind Maps give me a bird's-eye view I could never build manually

I let NotebookLM connect the dots for me

You know how when a murder mystery is being solved in a TV show or movie, the characters all huddle together to stick photos, sticky notes, and red string up on a corkboard, slowly connecting suspects to motives to evidence until the whole picture finally clicks? That's basically what NotebookLM's Mind Maps feature does for your notes, except you don't have to be the one stringing it all together.

Mind Maps are the workflow I lean on when a notebook gets too big to make sense of in one go. Audio Overviews are amazing when your notebook is focused on a single topic or week, but once you cross 15–20 sources, the podcasts start glossing over the connections between concepts, and that's where Mind Maps come in. NotebookLM scans everything in the notebook and generates a branching, clickable diagram of how your sources relate to each other — kind of like Obsidian's Graph View, except you didn't have to manually link a single thing.

My favorite use case for this is at the end of a semester, when I have an entire course's worth of materials sitting in one notebook like every week of lecture notes, slides, readings, supplementary YouTube videos I watched to make sense of harder topics, and my own messy scribbles. Generating a Mind Map gives me a literal map of the course! This is simply something you can't replicate with any other note-taking app, and it's why NotebookLM has quietly become the only place I bother dumping all my course material into.

Tweaking the notebook's personality turns it into a proper tutor

The one setting that does 80% of the heavy lifting

Over the months, I've seen a lot of people try NotebookLM for the first time. Here's how they go about it: they create a notebook, dump a source or two in it, and begin querying their notebook. That's....it. If you're leaving it at that, trust me when I tell you that you're using NotebookLM at maybe 30% of its actual potential.

NotebookLM has one setting that lets you customize the notebook's personality, and basically give the chat panel a role to play. You can tell it to behave like a strict examiner, a Socratic tutor that never gives you a direct answer, a friendly study buddy who explains things in plain English, or anything else you can dream up. I even turned it into a tutor that could only speak Subway Surfer brainrot once!

And because it's per-notebook, you can shape each one around the subject. For my theory-heavy courses, I've set the personality to never spoon-feed me an answer. It has to ask me a leading question first, which forces active recall every time I open the chat. For my exam prep notebooks, I tell it to act like a strict examiner: every time I attempt an answer, it grades my understanding and pushes me to defend my reasoning. This is one setting that transforms the Chat panel from a basic AI summarizer into something that actually teaches the way you want to learn, and it takes seconds to set up.

You're missing out if you're still treating NotebookLM like a chatbot

I've said this countless of times before and I'll say it again: you're missing out if you're treating NotebookLM like a glorified chatbot. We have tons of coverage up at XDA that goes deeper into the workflows I've covered here, plus a whole lot more!