How to Turn a Draft and Research Packet Into an Editor Memo With NotebookLM

The problem this solves and who it is for

This workflow is for editors and writers who already have a draft but do not trust it yet. The piece might read smoothly while still being thin, weakly supported, out of balance, or missing important context. A draft plus research packet is the perfect moment for a source-grounded editor memo.

The output is not a rewritten article. It is a memo that tells you where the draft is strong, where support is weak, where context is missing, and what the writer should fix next.

Prerequisites

A Google account with access to NotebookLM
A draft article in PDF, doc, or copied text
A research packet containing reports, notes, transcripts, or source pages
A clear understanding of the intended reader and article goal

How to capture or gather the source material

Export the current draft into PDF or keep it in a clean document.
Gather the source materials the draft is supposed to rely on.
Remove unrelated research from the packet. Only include what the draft should actually use.
Add one short note that states the article goal and target reader.
Name the draft file clearly so NotebookLM can distinguish it from the support sources.

Step-by-step workflow

Create a notebook for the article. Upload the draft, the research packet, and the article-goal note.
Ask for draft-versus-source comparison first. Have NotebookLM identify where the draft reflects the packet well and where it outruns the evidence.
Ask for weak sections second. Request a list of thin, unsupported, unclear, or context-poor sections.
Ask for an editor memo third. The memo should tell the writer what to fix first, what additional sourcing is needed, and which parts may need cutting or restructuring.
Revise outside the notebook. Use the memo to update the draft in your normal writing environment.

Tool-specific instructions

Primary recommendation: NotebookLM

NotebookLM is the best fit because the memo should be grounded in the draft and the supporting packet together. Google documents uploaded sources, notes, and source-based chat in NotebookLM, which makes it well suited for this kind of editorial comparison.

Practical setup:

Upload the draft as one source and the packet as separate supporting sources.
Ask specifically for places where the draft goes beyond what the sources support.
Save the final memo as a note so it stays attached to the source packet.

Alternative: ChatGPT Projects

ChatGPT can do this if you upload the draft and packet into a Project. It works well, but NotebookLM is usually easier to trust when you want the analysis closely tied to the sources.

Alternative: Claude

Claude is a good alternative if you want a direct editorial memo style. Upload the same packet and ask it to separate evidence gaps from structural issues.

Copy and paste prompt blocks tailored to the workflow

Draft-versus-source comparison prompt

{
  "role": "editor",
  "task": "compare a draft article against its uploaded research packet",
  "goal": "identify where the draft is strong, weak, unsupported, or missing context",
  "instructions": [
    "Use only the uploaded notebook sources.",
    "Treat the draft article as the item under review and the rest of the uploaded material as the support packet.",
    "Identify where the draft is well grounded, where it goes beyond the sources, where context is missing, and where stronger examples or proof are needed.",
    "Do not rewrite the article yet."
  ],
  "output_format": {
    "well_grounded_sections": [],
    "unsupported_or_overstated_sections": [],
    "missing_context": [],
    "areas_needing_stronger_examples_or_proof": []
  }
}

Editor memo prompt

{
  "role": "assigning editor",
  "task": "write an editor memo from a draft and research packet",
  "goal": "give the writer a clean revision plan",
  "instructions": [
    "Use the uploaded notebook sources and prior comparison analysis.",
    "Write an editor memo that includes what is working, what is weak, what is unsupported, what context is missing, what should be revised first, and what extra sourcing or reporting is still needed.",
    "Keep the memo practical and prioritized."
  ],
  "output_format": {
    "what_is_working": [],
    "what_is_weak": [],
    "what_is_unsupported": [],
    "missing_context": [],
    "revision_priorities": [],
    "additional_sourcing_or_reporting_needed": []
  }
}

Quality checks

Make sure the memo compares the draft against the packet, not against generic writing standards alone.
Confirm unsupported claims are named clearly.
Check that the memo prioritizes fixes instead of producing a long unfocused list.
Remove any criticism that is not grounded in the actual uploaded sources.

Common failure modes and fixes

Failure mode: The memo is too soft.
Fix: Ask explicitly for unsupported claims and weak evidence.

Failure mode: The memo is too long to use.
Fix: Ask for the top five revision priorities first.

Failure mode: The packet is bloated.
Fix: Remove background material the writer is not actually using.

Failure mode: The tool starts rewriting instead of evaluating.
Fix: Keep the request focused on memo creation, not draft improvement.

Sources Checked

https://support.google.com/notebooklm/answer/16164461 (accessed 2026-03-25)
https://support.google.com/notebooklm/answer/16215270 (accessed 2026-03-25)
https://support.google.com/notebooklm/answer/16262519 (accessed 2026-03-25)
https://help.openai.com/en/articles/10169521-projects-in-chatgpt (accessed 2026-03-25)
https://support.anthropic.com/en/articles/8241126-uploading-files-to-claude (accessed 2026-03-25)

Quarterly Refresh Flag

Review by 2026-06-23 to confirm tool interfaces and supported file workflows still match the live products.