Google recommends these 7 NotebookLM prompts, and they showed me what I was missing

Published Aug 13, 2026, 7:30 AM EDT Mahnoor Faisal is a tech journalist covering AI and productivity tools with bylines at XDA, SlashGear, MakeUseOf, Laptop Mag, and Android Police. She's been writing professionally since she was sixteen, and has since penned hundreds of articles. This includes in-depth coverage of AI tools like NotebookLM to breaking news across the AI space. Her passion for technology started when she received her first iPod Touch (4th generation) on her 8th birthday, and she's been deep in the tech world ever since. Currently pursuing a degree in computer science, Mahnoor brings both a journalist's eye and a technical foundation to her coverage of how AI is reshaping the way we work and learn. If there’s one group of people I’m willing to take prompting advice from, it’s the people actually building the AI tool I’m using. I’ve spent plenty of time stealing tips from Claude Code’s creator and Anthropic’s own team, and it’s become one of my favorite ways to figure out what these tools are actually capable of. I’ve even sat down with a senior engineer on the NotebookLM (now Gemini Notebook) team to learn how he uses the tool himself, and walked away with insights I hadn’t considered even after writing more than 200 articles about it. So, when Google recently shared seven prompts it recommends trying with NotebookLM’s latest capabilities, I obviously had to put them to the test. Students still get the most obvious wins Even the student prompts had a few surprises As a student, NotebookLM has always been the first AI tool I reach for when I’m studying. There are two reasons I’ve come to rely on it so much. The first is its grounded approach. Rather than pulling an answer out of the wider internet by default, NotebookLM has traditionally worked from the sources I give it and points me back to those sources with citations. That makes it much easier to trust when I’m working with lecture notes, readings, or material I actually need to learn. The second is that NotebookLM’s student-focused features have always felt genuinely practical. The recent updates, though, have left me largely disappointed. I've struggled to see how some of them fit into the way students actually use NotebookLM. Some of the updates also make the tool feel like it's drifting away from the focused, source-grounded experience that made it so useful for students in the first place. That's why I was especially curious to see the three prompts Google itself recommends for learning. The first two prompts Google recommends perfectly illustrates what's different about the "new" NotebookLM. Create 5 quizzes of ascending difficulty focusing on a different topic from my sources Turn my raw lecture notes into a formatted PDF study guide complete with a 5-question practice quiz. Neither quizzes nor study guides are particularly new territory for NotebookLM. What is new is that I no longer have to think nearly as much about how to create them. Instead of opening Studio, picking an artifact, adjusting its settings, and repeating the process for every variation I want, I can now describe the finished result in natural language and let NotebookLM handle the steps in between. That sounds like a small change, but the first prompt makes the difference especially obvious. Asking for five quizzes of ascending difficulty, each focused on a different topic, would previously have felt like a series of separate tasks. Now, I can spell out the whole thing in a single sentence and let NotebookLM build the artifacts for me. The second prompt takes the same idea even further. Rather than asking NotebookLM to summarize a set of messy lecture notes and then separately turning that summary into something useful, Google suggests asking for the final product outright: a formatted PDF study guide, complete with a practice quiz. The third prompt, though, is what I find most handy as a student. I spend a good chunk of time manually adding sources to NotebookLM, and when a course syllabus links out to readings, papers, and other material, that can quickly turn into a tedious copy-and-paste exercise. Add all of the URLs from this class syllabus as sources This prompt Google recommends basically skips that entire process. I can upload the syllabus, tell NotebookLM what I want, and have it find the links and build out the notebook for me. This is something NotebookLM couldn't previously do, but it's an extension of the same source-first workflow that made the tool so useful to me in the first place. The business prompts are where NotebookLM starts doing the work NotebookLM has entered its analyst era I don’t own a business, nor does my job involve working directly with businesses, so these next two prompts admittedly aren’t ones I’d naturally reach for. Still, the first one might be the clearest demonstration yet of how different the new NotebookLM really is. Calculate revenue growth across each product line and plot the trends. Previously, NotebookLM was largely a tool that could read the information you give it, help you make sense of it, and surface connections you might have missed. Now, it can actually take that data and do something with it. As part of its recent upgrade, NotebookLM gained the ability to run code in a secure cloud environment, meaning it can perform calculations and data analysis itself, then turn those results into charts and other downloadable files. Google’s second business prompt pushes that shift in a slightly different direction: Read our marketing brief, search the web for competitors, and recommend how we should tune our messaging. This one combines two versions of NotebookLM that used to feel much more separate: the source-grounded tool I’ve always known and the newer, more agentic one Google is building. It starts with a document you’ve provided, but it no longer has to stay inside that closed collection of sources. It can head out to the web, gather additional information, compare what it finds against your own material, and use all of that to recommend what you should do next. The personal prompts push NotebookLM into everyday life My notebooks are getting suspiciously practical The personal prompts are probably the ones I found most interesting, because they make NotebookLM feel less like a research tool and more like something I could hand an annoying real-life task to. Google’s first suggestion is: I uploaded a bunch of receipts from our kitchen renovation, make a spreadsheet to track all the work and costs. I’m not renovating a kitchen anytime soon, but I immediately understood the appeal. Receipts are exactly the kind of source material I’d normally upload to NotebookLM to ask questions about later. The difference now is that I can skip the question-and-answer stage entirely and ask it to turn that messy pile of information into something useful. Instead of reading through every receipt myself, pulling out vendor names, dates, amounts, and what each payment was for, then manually entering everything into a spreadsheet, NotebookLM can handle that transformation for me. It’s another example of the shift I kept noticing throughout Google’s recommended prompts — the sources are no longer necessarily the end point. They can just be the raw material NotebookLM uses to create whatever I actually need! The final prompt takes that idea much further: Help me analyze this property purchase. Look up recent market performance in the neighborhood, predict trends, and make a spreadsheet to model all the data you find. This is easily the most ambitious of Google’s seven suggestions. Instead of just asking NotebookLM to understand something you uploaded. You’re asking it to take your own information, head out to the web for additional research, analyze what it finds, make projections from that data, and package everything into a spreadsheet you can actually work with. I still don’t love every direction NotebookLM has taken recently, but Google’s prompts helped me understand what it’s trying to become. So, credit where credit is due. The update makes a lot more sense once you see what Google actually expects people to do with it.