How to Turn Showing Feedback From Multiple Sources Into a Pattern Summary With NotebookLM
Collect showing feedback from several channels into a source-grounded, frequency-checked pattern summary for listing decisions.
Workflow summary
Quote this workflow accurately
- Best for
- Showing comments rarely arrive in one tidy place. Buyers' agents email reactions, open house hosts keep notes, and listing teams receive texts. For agents and property operations staff, the task is to distinguish repe...
- Input
- A Google account authorized to use Gemini Notebook with the information involved. The showing feedback emails, text excerpts, open house comments, and CRM notes for one property . The listing description and enough co...
- Primary tool
- NotebookLM
- Output
- Collect showing feedback from several channels into a source-grounded, frequency-checked pattern summary for listing decisions.
- Main risk
- One memorable comment dominates. Ask for a frequency table with a list of distinct source IDs and manually verify it.
- Verification step
- Every claimed repeated theme maps to two or more distinct, real feedback events, not duplicated forwards. Opinions are attributed to the correct party and separated from facts about the property. Summary language avoi...
Continue from this article
Problem and who this is for
Showing comments rarely arrive in one tidy place. Buyers' agents email reactions, open-house hosts keep notes, and listing teams receive texts. For agents and property-operations staff, the task is to distinguish repeated property objections from a memorable but isolated remark, then draft a concise seller update.
Google renamed NotebookLM to Gemini Notebook on July 16, 2026. The original article title and slug are retained for existing search and internal links; current instructions use Gemini Notebook. Responses are source-grounded but AI-generated, so citations are starting points for checking, not proof of accuracy.
Prerequisites
- A Google account authorized to use Gemini Notebook with the information involved.
- The showing feedback emails, text excerpts, open-house comments, and CRM notes for one property.
- The listing description and enough context to interpret comments without exposing unnecessary client details.
- A person responsible for checking the final seller-facing report.
Before uploading, remove unnecessary personal information and confirm that the account, source permissions and retention practices are appropriate. A notebook viewer can access source documents even if the shared interface initially hides them. Do not share a source notebook as the finished client-facing deliverable.
How to capture or gather the source material
Create a dated feedback log. Each row should identify the showing or open-house event, channel, who is speaking (buyer, agent, or listing team), verbatim comment if available, and a source reference. Use a unique event ID so two agents repeating the same buyer's opinion are not counted as two independent objections. A Google Doc, CSV, PDF, or several clean text files all work as notebook sources. Remove phone numbers, buyer names, access codes, and unrelated personal information.
Step-by-step workflow
- Create a listing-specific notebook. On the Gemini Notebook website, select Create new notebook and add the source log, listing description, and any supporting notes. Do not mix separate listings.
- Check source coverage. Ask which feedback records are present, which dates are missing, and whether one comment appears in more than one file. Repair the log before analyzing patterns.
- Request a grounded classification. Paste the JSON prompt below into Chat. Ask the tool for cited examples and data-quality flags, not a decisive pricing recommendation.
- Validate actual frequency. Count independent showing events behind each recurring theme in the original log. A phrase repeated in three forwarded email copies is still one comment.
- Make a short action memo. Keep no more than three next-step questions, such as revising confusing listing photos or checking whether buyers misunderstood a floor-plan feature. Pricing, marketing and representation decisions remain with the licensed professional.
- Create the seller version. Remove identifying source details, confirm each reported pattern, and share a clean document or report, not the notebook containing raw messages.
Tool-specific instructions
Gemini Notebook uses the imported sources as the basis for Chat responses. The Studio panel can create reports, infographics and slide decks if the seller would benefit from a visual summary. A report exported to Google Docs is a separate artifact; edits in that file do not flow back to the notebook. For accurate counts, work from unique feedback rows in a spreadsheet and compare the model's grouping to those rows. Do not assume AI can distinguish duplicate events automatically.
Alternative: If source messages cannot be uploaded under your brokerage policy, use a sanitized feedback matrix in the firm's approved document or spreadsheet tool and write the seller memo from the checked patterns.
Copy/paste prompt
{
"task": "Extract evidence-backed feedback patterns for one property listing.",
"input_scope": "Only the selected listing feedback logs, messages and property context sources.",
"instructions": [
"Separate direct buyer statements, buyer-agent paraphrases and listing-agent interpretation.",
"Group comments into positives, objections, questions and one-off remarks.",
"Give exact counts only when individually identifiable feedback records support them; report unique commenting parties as well as comments if possible.",
"Include brief citations or exact source identifiers for each pattern.",
"Do not infer buyers’ personal characteristics or propose demographic targeting.",
"Keep recommendations as possible questions for the listing agent, not automated pricing decisions.",
"Mark missing dates, duplicates and conflicting records explicitly."
],
"output_format": {
"source_inventory": [],
"repeated_positives": [],
"repeated_objections": [],
"one_off_comments": [],
"open_questions": [],
"possible_actions_max_3": [],
"data_quality_warnings": []
}
}
Quality checks
- Every claimed repeated theme maps to two or more distinct, real feedback events, not duplicated forwards.
- Opinions are attributed to the correct party and separated from facts about the property.
- Summary language avoids demographic stereotypes or assumptions about buyers.
- No buyer contact information, access instruction, negotiation detail, or other restricted material appears in the seller report.
- Proposed actions are proportionate to the volume and quality of feedback.
Common failure modes and fixes
One memorable comment dominates. Ask for a frequency table with a list of distinct source IDs and manually verify it.
A model reports made-up percentages. Remove percentages and use exact verified counts from the underlying feedback register.
A repeated complaint is actually one forwarded message. Deduplicate by showing event and source before summarizing.
Recommendations become definitive pricing advice. Keep the AI output as documented feedback and possible follow-up questions for the agent to evaluate.
Tools Used
Gemini Notebook, formerly NotebookLM: /go/notebooklm. Review the site's affiliate disclosure before activating any affiliate redirect.
Sources Checked
- Google, NotebookLM is now Gemini Notebook, July 16, 2026: https://blog.google/innovation-and-ai/products/gemini-notebook/notebooklm-gemini-notebook/ (accessed 2026-10-14).
- Google, Add or discover new sources for your notebook: https://support.google.com/gemininotebook/answer/16215270?co=GENIE.Platform%3DDesktop&hl=en (accessed 2026-10-14).
- Google, Create a notebook in Gemini Notebook: https://support.google.com/gemininotebook/answer/16206563?hl=en (accessed 2026-10-14).
- Google, Share and manage notebooks: https://support.google.com/gemininotebook/answer/16206563?co=GENIE.Platform%3DDesktop&hl=en (accessed 2026-10-14).
- Google, Generate reports in Gemini Notebook: https://support.google.com/gemininotebook/answer/18323649?co=GENIE.Platform%3DDesktop&hl=en (accessed 2026-10-14).
- Federal Trade Commission, Protecting Personal Information: A Guide for Business: https://www.ftc.gov/business-guidance/resources/protecting-personal-information-guide-business-0 (accessed 2026-10-14).
Quarterly Refresh Flag
Review by 2027-01-12. Revalidate the current product name, source-import options, account-sharing settings and workflow-specific instructions before republishing.
Related Workflows
How to Turn a Post-Showing Voice Memo Into a Seller Recap With NotebookLM
Record a quick voice memo after a showing, upload it with your listing context, and turn it into a clean seller recap with reactions, objections, follow-up questions, and next steps.
How to Turn Handwritten Showing Notes Into a CRM-Ready Recap With AI
Snap a photo of handwritten notes after a property tour or showing block and convert them into a structured recap with reactions, priorities, and follow-up tasks you can paste into your CRM.
How to Turn Vacant Room Photos Into a Quick Virtual Staging Concept Deck With AI
Use Gemini to stage multiple vacant room photos and turn the strongest outputs into a quick concept deck for owner, seller, or investor review.
Help keep PromptedWork sharp
Share a broken step, outdated prompt, or general feedback. This is only for improving this specific workflow.