The whole play, at a glance
You walk away with: a pricing narrative a seller with no real estate background can follow, two-sentence comp cards, honest caveats that build trust, and grounded responses to the three objections this specific home will raise.
What you'll need
Comps from your MLS or CMA tool. Export the summary, or just copy the comp grid: sold prices, dates, size, condition notes.
Your AI Second Brain workspace.
The setup
The AI never picks comps and never picks the price. It builds the story around the comps you chose.
Paste the comp grid into your workspace, or upload the CMA summary export.
Does the story match what you know about the market? Fix anything that doesn't ring true before it goes near a seller.
The shape of the output: a story section, comp cards, and honest caveats. Your packet's narrative pages, drafted.
The AI plays the seller and gives you the three hardest pricing objections for this specific home, phrased the way they'll actually come across the kitchen table.
AI transcription of tables is good, not perfect. One transposed sold price costs more credibility than the whole CMA earned.
The prompts
The Pricing StoryComps in, plain-English narrative out
You are my listing-presentation assistant. Below are the comps I selected for a CMA (comparative market analysis). I will decide the price; your job is the story. SUBJECT PROPERTY: [address/area, size, condition, standout features] MY COMPS: [paste the comp grid: sold/active/pending, prices, dates, size, condition notes, adjustments if you have them] Produce: 1. THE STORY: 3-4 short paragraphs a seller with no real estate background would understand. What the comps show, why these homes are the right comparisons, what the market is doing right now in this area, and what range the data supports. Plain English, no jargon without a translation. 2. COMP CARDS: for each comp, two sentences: why it's relevant, and what it tells us. 3. THE HONEST CAVEATS: what this data can't tell us (thin comp set, fast-moving market, unique features), stated in a way that builds trust instead of doubt. Use only the data I pasted. If something looks inconsistent, flag it instead of smoothing it over.
Objection PrepThe AI plays the seller; you get grounded responses
Based on the subject property and comps above, play the seller. Give me the 3 hardest pricing objections this specific seller is likely to raise (for example: the online estimate is higher, the neighbor got more last year, we did upgrades). For each objection: how it will actually be phrased across the kitchen table, and a response that is honest, grounded in the comp data, and keeps the relationship warm. No scripts that sound like scripts.
Watch-outs
You price the home. "AI-powered pricing" is a claim you can't stand behind and shouldn't make. AI explains; you and the data decide.
MLS data has rules. Using comps with your client is normal course of business; republishing raw MLS data publicly is not. Keep the comp detail in the appointment.
Verify before the client sees it. Every price, date, and square footage gets checked against the MLS.