Descriptions:
Parker Prompts shares a systematic approach to configuring Google’s NotebookLM for research and knowledge work, drawing on 3 months and more than 80 notebooks of hands-on experience. The central argument is that NotebookLM’s out-of-the-box behavior — generic summaries that any chatbot could produce — is a configuration problem, not a capability problem.
The most impactful fix is the ‘Configure Chat’ panel in the top-right corner of the interface, which most users never touch. Setting a specific role and conversational goal (rather than leaving it on ‘Default’) shifts the model from broad summarization to targeted, expert-style analysis. The video demonstrates this directly: the same question on the same sources produces noticeably more actionable output with a custom role active. The response length toggle — set to ‘Longer’ for deep research, ‘Shorter’ for daily summaries — is a second quick win that changes the character of every response in a notebook.
The standout section covers Audio Overview format options that go almost entirely undiscovered. Beyond the default ‘Deep Dive’ format, NotebookLM offers Brief, Critique, and Debate modes. The Critique format is particularly distinctive: instead of summarizing, the AI hosts actively identify weaknesses, gaps, and unsupported claims in the uploaded material — making it useful for stress-testing a business plan, research proposal, or piece of writing before external review. The video also covers NotebookLM’s built-in Deep Research tool, including a prompting approach that filters results to academic papers and industry data while excluding opinion content.
📺 Source: Parker Prompts · Published March 23, 2026
🏷️ Format: Tutorial Demo
