Redesigning a unit with existing teaching materials with NotebookLM

The challenge I faced was not deciding what to teach; the materials already existed. The harder question was how to teach it effectively: identifying where student confusion typically arises, ensuring content aligns with unit learning outcomes, updating activities, and reorganising the unit's flow across the semester. Curriculum redesign from a large, unstructured bank of legacy materials is time-consuming work, and NotebookLM offered a way to do it more systematically.

With all the previous materials uploaded, I began by asking NotebookLM to identify the most important things students needed to know, using the prompt:

"What are the top 10 things students need to know according to everything that I've written?" 

From there, I asked it to map those priorities against the unit's learning outcomes, which surfaced gaps and misalignments I could then address.

Then I moved into content generation. Using prompts like "give me a worksheet" and "make slides from that worksheet", I produced draft teaching resources directly from existing material. When I didn't like a generated diagram, I simply prompted: "I don't like the diagram, make another one".

For semester planning, I asked: "I've got 8 weeks in this unit, give me an 8-week plan so I can cover all these things".

I also used prompts to help contextualise content for students: "Here are the topics, how can I explain to my students how these could all be used in practice?" and "give me some research questions based on these topics".

Watch the video below to see the NotebookLM outputs and what they looked like

This use case demonstrates that NotebookLM can be particularly valuable for curriculum review and redesign when educators already hold a substantial bank of legacy materials. Rather than starting from scratch, it allows you to work systematically with what you already have, surface hidden gaps, reveal where students may be confused, and accelerate the early stages of design work. The final polishing or more creative tasks may still happen in other tools, but the groundwork can be done significantly faster.

I found the overall process genuinely useful for interrogating and reorganising my existing materials. Though there were several limitations worth noting.

• The automatically generated slides felt overproduced for my purposes.
• The source limit became a practical constraint. I could not add more materials once I reached 49 sources.
• For some tasks, particularly generating realistic practice applications or producing high-quality research questions, I found Gemini performed better than NotebookLM.
• I have begun introducing students to NotebookLM through reading lists, though formal student feedback has not yet been collected.