thinkhard.SI

Prompt repository Small business

Turn a home buyer's reactions to showings into a sharper search

You have shown a buyer several homes and want to turn what they liked and disliked into clearer criteria and a better next round, within fair housing rules.

Registry no.
SMB-070
Collection
Small business
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
4

Small businessAny chat

The prompt

Score 10 of 10, Excellent

Use it when You have shown a buyer several homes and want to turn what they liked and disliked into clearer criteria and a better next round, within fair housing rules.

  • {{buyer}}
  • {{original_criteria}}
  • {{showing_notes}}
  • {{current_search}}
Full prompt and why it works
Prompt
Help me turn my buyer's reactions to the homes we have seen into a sharper search, for my next conversation with {{buyer}} and for setting up their listing alerts. Buyers often say they want one thing and react to another, so the useful work is spotting the pattern in what they actually liked and rejected, then checking it with them rather than deciding for them.

What the buyer told me they wanted at the start (budget range from their lender, must-haves, nice-to-haves, timing): {{original_criteria}}

The homes we have seen, with their facts and what the buyer said or did at each, as close to their words as I noted:
"""
{{showing_notes}}
"""

The search settings I use now: {{current_search}}

Produce, with a heading for each part:
1. Patterns: what the buyer's reactions show, each tied to the showings it comes from, for example "Rejected three homes with small kitchens, even when the price was right (homes 2, 4 and 5)." That example shows the form only.
2. Where reactions and stated criteria disagree, as questions I can ask, such as "You said a yard was a must, but you liked the condo on [street]. Is the yard still a must?"
3. Revised criteria: must-haves, nice-to-haves, trade-offs they seem willing to make and deal-breakers, each marked "confirmed" or "to check with the buyer".
4. Suggested search changes: filters and features to add or drop, with the price range and areas only as the buyer and their lender have set them.
5. A short message to the buyer, under 120 words, summarizing what I noticed and inviting them to correct me.

Fair housing rules for every part: describe homes and features, never the people in a neighborhood. Do not suggest or rule out areas based on who lives there, schools, safety or other judgments about a community; if the buyer raised these, note that they can check public sources themselves. Do not judge what the buyer can afford beyond what their lender set.

A good result is a set of criteria the buyer reads and says "yes, that's what I want". Use only my notes. If a showing note is too thin to read a reaction from, say so instead of guessing what the buyer felt.

Variables

{{buyer}}
Your buyer client, for example "a couple relocating for work".
{{original_criteria}}
What they asked for at the start: budget range from their lender, must-haves, nice-to-haves and timing.
{{showing_notes}}
Each home you showed, its facts, and what the buyer said or did there.
{{current_search}}
Your current search filters, for example "3 bedrooms, under the lender's limit, two named areas".

Why it works

  • Background facts. Showing notes close to the buyer's own words let the AI find patterns in what they did, not only in what they said at the first meeting.
  • What to avoid. Fair housing rules apply to every part: homes and features are described, never the people in an area, and affordability stays with the lender.
  • If unsure. Each revised criterion is marked confirmed or to check, and a note too thin to read is flagged instead of interpreted.

How to use it

  1. Press Fill in blanks and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
  4. When the answer comes back, check it before you use it. The Fact Check List lists every figure, quote and source in the answer worth checking, riskiest first.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/buyer-showing-feedback-search-brief

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "buyer-showing-feedback-search-brief"}, or just ask for it by name.