Most real estate agents who try ChatGPT for the first time type something like "write a listing description for a 3-bedroom house in Austin" and get back generic filler that sounds like every other listing on Zillow. They conclude AI isn't useful and go back to doing everything by hand. That's the wrong conclusion — the problem is the prompt, not the tool.
The agents actually saving time with AI — 6 to 10 hours a week in many cases — have built a library of specific, structured prompts that encode the context the AI needs to produce professional-grade output. This guide gives you those prompts, explains why they work, and shows you how to adapt them to your market and your voice.
We work with real estate agents and teams on AI implementation, so what follows comes from real production use — not theoretical prompts that look good in a blog post but fall apart in practice.
Why Most Real Estate AI Prompts Fail
A prompt is only as good as the context you give it. ChatGPT and Claude are language models — they generate the most probable continuation of what you've given them. If you give them sparse input, they fill in the gaps with generic, statistically average content. That's where "granite countertops and stainless steel appliances" comes from.
Good prompts do three things: they tell the AI who the audience is, what specific details to use, and what tone and style to match. The prompts below are built on that framework. You'll notice they're longer than what most agents try — that's intentional. More specific input reliably produces better output.
Rule of thumb: The time you spend writing a detailed prompt is returned 5x in the quality of what you get back. A 2-minute prompt setup saves 20 minutes of editing a mediocre draft.
Listing Description Prompts That Actually Sell
The goal of a listing description isn't to describe the house — it's to help the right buyer picture themselves living in it. Here's a prompt structure that produces descriptions worth publishing:
// Listing Description PromptWrite a compelling MLS listing description for the following property. Audience: [first-time buyers / move-up buyers / investors / downsizers — pick one]. Tone: warm and aspirational, not corporate or listy. Length: 180–220 words. Do not use the phrases "stunning," "must-see," "won't last long," or "motivated seller."
Property details:
- Address/area: [neighborhood, cross streets, or city]
- Bedrooms/baths: [X bed / X bath]
- Square footage: [X sqft]
- Lot size: [X sqft / acres]
- Year built: [XXXX]
- Key features: [list 6–10 specific features — be precise, e.g. "10-ft ceilings in the main living area" not just "high ceilings"]
- Recent updates: [list anything updated in last 5 years with approximate year]
- Neighborhood highlights: [walkable to X, near Y school, Z-minute commute to downtown, etc.]
- One thing that makes this home unusual or memorable: [be specific]
Write two versions: one that leads with lifestyle, one that leads with the home's most distinctive physical feature.
The "two versions" instruction is worth keeping. It forces the AI to approach the property from different angles, and you'll often find the second version is stronger — or you'll combine elements from both.
Buyer Follow-Up Sequences
A buyer views a property, doesn't make an offer, and goes quiet. What do you send? Most agents either send nothing or a generic "just checking in" email that gets ignored. Here's a prompt that generates a real follow-up sequence:
// Buyer Follow-Up Sequence PromptWrite a 3-email follow-up sequence for a buyer who toured a property but hasn't made an offer. The emails should be spaced: Day 2, Day 5, Day 10 after the showing.
Buyer profile: [describe briefly — e.g., "young couple, first home, budget around $450K, prioritize good school district, seem nervous about the market"]
Property they toured: [address or brief description]
Why they may be hesitating: [e.g., "price is at the top of their budget" / "comparing to two other homes" / "not sure about the neighborhood"]
Email 1 (Day 2): Warm check-in, address their main hesitation directly, offer one specific piece of value (a data point, a resource, a question answered).
Email 2 (Day 5): New angle — share a market insight or comparable sale relevant to their decision. No pressure, but create mild urgency if the market warrants it.
Email 3 (Day 10): Final value-add touch. Offer to answer questions, suggest alternatives if this home isn't right, keep the relationship warm regardless of outcome.
Tone: Professional but human. This agent is a trusted advisor, not a salesperson pushing a close. Subject lines included for each email.
Market Update Emails for Your Database
Agents who stay top-of-mind with their sphere close more referral business. Monthly market updates are one of the most effective ways to do this — but writing them consistently is hard to sustain. Here's a prompt that makes it fast:
// Monthly Market Update Email PromptWrite a monthly market update email for my real estate database. This is not a newsletter — it's a short, direct market snapshot from a local expert. Length: 250–300 words maximum. No filler, no generic encouragement.
My market: [city/area]
Month: [Month Year]
Key stats to include:
- Median sale price: [$X, up/down X% from last month / last year]
- Days on market: [X days average]
- List-to-sale price ratio: [X%]
- Active inventory: [X homes, up/down X% from last month]
- Notable neighborhood or price segment trend: [one specific observation]
My take: [1–2 sentences of my personal interpretation — what does this mean for buyers? For sellers?]
Call to action: Invite reply with questions, not a hard sell.
Subject line options: give me 3 subject line options, one data-led, one curiosity-led, one conversational.
A solo agent in Scottsdale sends this email to 1,200 past clients and sphere contacts every month. She estimates it generates 2–3 warm conversations per send that wouldn't happen otherwise. At her average commission, that's a substantial annual return on a 15-minute monthly task.
Offer Presentation Scripts
When you're presenting an offer on behalf of a buyer in a competitive situation, how the offer is framed matters. Many agents write a brief cover letter to accompany the offer — but most of these letters are weak because they're written under deadline pressure. This prompt helps:
// Offer Cover Letter PromptWrite a professional offer cover letter from a buyer to a seller. This letter accompanies a purchase offer and is meant to be read by the listing agent and seller. It should be genuine and compelling without being manipulative or overly emotional. Length: 150–200 words.
Buyer profile (share only what's appropriate and relevant): [e.g., "young family relocating from Seattle for work, two kids, want to put down roots"]
What they love about this specific property: [list 2–3 genuine, specific things — not generic praise]
Offer strength: [e.g., "full asking price, 21-day close, no inspection contingency" — only include real terms]
Any personal connection to the neighborhood or home: [optional but powerful if genuine]
Tone: Warm, confident, and human. Avoid anything that sounds like a negotiating tactic or that would feel off-putting to a seller who is emotionally attached to their home.
Social Media Content in Bulk
Consistency on social media builds brand presence for real estate agents — but creating content every week is genuinely hard to sustain alongside a full transaction load. Here's how to batch-create a month of content in one session:
// Monthly Social Content Batch PromptCreate 12 social media post captions for a real estate agent. These will be posted 3x per week across Instagram and Facebook. Mix the following content types (4 posts each): market insight, lifestyle/neighborhood, educational tip for buyers or sellers.
Agent's market: [city/area]
Agent's personality/voice: [e.g., "direct, data-driven, a little dry humor — not cheerful or corporate"]
Current market condition in one sentence: [e.g., "inventory is tight, prices holding, buyers need to move fast"]
Recent listings or sales to reference: [optional — include 1–2 if you want to weave in social proof]
For each post: write the caption (150 words max), suggest a content format (photo, reel, carousel, graphic), and suggest a hashtag set of 5–8 relevant tags. Number each post 1–12.
Batching is the key: Agents who try to create one social post at a time burn out quickly. Set aside 90 minutes once a month to generate a full month of content using a single prompt session. Review, lightly edit, schedule. Done.
CMA Narrative Summaries
Comparative Market Analyses are data-heavy and often impenetrable to clients. Agents who add a plain-English narrative summary — explaining what the comps mean and what pricing strategy they recommend — close more listings. Here's a prompt for that summary:
// CMA Narrative Summary PromptWrite a plain-English summary of a comparative market analysis for a seller consultation. This will be read by a homeowner who is not a real estate expert. Tone: clear, confident, like a trusted advisor explaining their recommendation — not a report.
Subject property: [brief description]
Comparable sales (include 3–5):
- [Address, sqft, beds/baths, sale price, days on market, sale date, key differences from subject]
Active competition:
- [List 1–3 active listings competing for the same buyer pool]
Key market context: [1–2 sentences on current market conditions]
Recommended price range: [$X–$Y]
My rationale for that range: [explain your thinking in 2–3 sentences]
Output: a 300–400 word narrative that walks the seller through the data, explains the recommended range, and pre-handles the most common objection (pricing too high to "leave room to negotiate").
Building Your Own Prompt Library
The prompts above are starting points. The real asset is a personal prompt library tailored to your voice, your market, and your specific client types. As you use these prompts, you'll find yourself making the same tweaks every time — that's a signal to bake those adjustments into the base prompt.
Agents who've gone through our real estate AI setup process end up with 20–30 prompts covering every common writing task in their practice. New team members can use them immediately. Consistency goes up. Time spent on admin goes down.
If you want to skip the trial-and-error and get a custom prompt library built for your specific business — along with training on how to use it efficiently — that's exactly what our AI Starter package covers. See the full breakdown of what's included in our pricing section.
Get a Custom AI Prompt Library for Your Real Estate Business
We'll build prompts matched to your voice, your market, and your client types — plus train your team on how to use them. Free evaluation, no commitment.
Get Your Free AI Evaluation
Real Estate AI Details