10 Hacks Every NotebookLM User Should Know

Google may spend more time promoting its Gemini AI chatbot, but NotebookLM (the LM stands for Language Model) is built on the same underlying AI, and it takes the analysis and deep dive features up a notch.

NotebookLM is built around the idea of a digital notebook, where various different sources are gathered together and scrutinized. Using data you import, NotebookLM can produce everything from video explainers, to podcasts, to presentation slides, to flashcard study aids from your selected sources.

It's an app packed with features, available on the web, on Android, and on iOS. If you've tried it out and you're ready to level up your NotebookLM experience beyond the basics, these 10 tips will take you there.

You can make NotebookLM to find its own sources

The traditional way to get started with NotebookLM is to feed it one or more sources—PDFs, web links, or YouTube videos for example—but the app can actually head out on the web and find its own relevant sources too.

If you want to research and explore a topic but are starting from scratch, this can be really useful. The Search the web box is in the Sources panel, and you can choose between Fast research and Deep research, depending on how quickly you want your results back.

When the results appear, select the ones you'd like to use and click Import. You can continue to look for new sources as needed, tweaking the information you're searching for each time.

You can be a "guest" on your Audio Overview podcasts

You may have played around with the Audio Overview feature in NotebookLM already, which lets you create realistic-sounding podcasts from your material, but there's a way to add your own interjections to these podcasts—like a caller ringing into a radio station.

Click the waving hand icon next to any Audio Overview to go into interactive mode, and when you've got something to say, click Join. The AI hosts will break to let you have your say, and then respond to your comment or question before resuming the podcast.

You can make NotebookLM show you overviews of individual sources without typing any prompts

Once you've uploaded all your materials into NotebookLM, you're probably going to want to busy yourself with asking questions about them, but you can make NotebookLM provide brief overviews of your sources without typing any prompts.

Select an individual source from the Sources pane and you'll see a Source guide up at the top: This is a handy AI-generated overview of the source and the content it contains. It will give you a good understanding of the material, and help you decide if it is a source you want to include in future queries. There are also tags down at the bottom of the guide that cover the key subjects in the source—click on any of these to get NotebookLM to tell you more about that subject in particular.

You can give each chat with NotebookLM its own personality

Up at the top of each Chat pane there's a configuration button (it looks like three sliders). Click this, then select Custom to get specific about how you want NotebookLM to respond in this conversation: You can ask it to play a certain role, tailor its output to a particular audience, or work towards a set goal.

For example, you can ask for replies to be aimed at a high school level of understanding, indicate that you're writing up a report for a board meeting, or get the AI to break down its responses into short, separated bullet points. The instructions you give it here get applied for the rest of the chat, until you change them again.

You can upload existing slideshows for NotebookLM to use as an example when making yours

NotebookLM is able to produce presentations based on your sources—either with a prompt or via the Slide deck option in the Studio panel—but it can sometimes be tricky to get these presentations looking exactly the way you want.

You can fix this by uploading an existing slideshow and using it as a reference point. Add it as a source, and then refer to the name of the file in your request for a new presentation: Just tell NotebookLM to use the existing slideshow as a style template, and it will.

NotebookLM's sharing options can your notebooks with the world

You don't have to keep your notebooks to yourself: While clicking the Share button on a notebook page will allow you to collaborate with other people on projects (and you get controls over how much they can edit and view), you have more options than that. If you select Anyone with a link under Notebook access, then click Copy link, you'll get a URL you can share with anyone, or post publicly on the web. (Here's one on the Made by Google 2025 event.)

This link will lead anyone who finds it directly to your notebook. They will get their own private chat history, but they won't be able to make any changes to sources or Studio materials.

You can use NotebookLM to sort your Google Drive

As you might expect, NotebookLM works well with other Google apps, and if you're comfortable giving it access to your files then the AI tool can work really well as a way of searching through your Google Drive and pulling out information from it.

When you click Add sources, you then need to click the Web drop-down menu and pick Drive instead. Enter something you're looking for, and you'll get a list of matches from your Drive storage: You can then use the checkboxes to decide which files to import.

Whether you're working on your next novel or trying to wrangle data from multiple spreadsheets for a business report, it makes it easy to mine the data inside whatever you've got stored in Google Drive.

You can tailor the sources for each prompt

You don't have to use every source for every prompt: Use the checkboxes next to each entry in the Sources panel to tell NotebookLM where to pull its information from (the sources overview mentioned above can be helpful here). There might well be times when you want to analyze every source except one, or you just want to focus on one specific source—maybe you have a bunch of reports you're analyzing and there's one in particular you want to interrogate, for example.

Use Google Docs to deliver more complex prompts

If you're working on really complex prompts—full of instructions, references, and multi-level queries—then these can work best as separate documents within your notebook. It means you've got more flexibility when composing and editing them, and once they have been made, you can refer to them again and again.

Make a new Google Docs document and then compose your prompt instructions as required—you can even split it up with headings and bullet points if needed. Then, add it as a source in NotebookLM.

For your next prompt, reference the name of the document and ask NotebookLM to use its contents as a framework for what it should do next. The AI will confirm that it's read and understood the document, and then proceeds as instructed.

Pull information from all of your notebooks at once by uploading them to Gemini

Your NotebookLM notebooks will also be available in the Google Gemini app. Select the + (plus) button next to the Gemini prompt box, and NotebookLM comes up as an option. You can then pick one or more notebooks to import. There are lots of ways that this can be useful, from using Gemini to create videos and images based on your collected material, to running prompts that analyze multiple notebooks at once (saving you having to switch between them in NotebookLM).