AI property valuation: 20 ChatGPT prompts to use today

2026
Article 09 · Prompts & Use Cases

AI property valuation:
20 ChatGPT prompts
to use today.

Florian Berthoud 11 min read Published July 2026
Aerial view of a real estate team working together on market data with laptops and notebooks, illustrating daily ChatGPT prompt use

Photo: Vlada Karpovich · Pexels

Article contents
  1. Before you start: the 3 rules of a good prompt
  2. Listing writing, 4 prompts
  3. Showing reports, 3 prompts
  4. Seller appointment prep, 3 prompts
  5. Buyer objection handling, 2 prompts
  6. Social media content, 3 prompts
  7. Document analysis, 3 prompts
  8. Email & prospecting, 2 prompts
  9. How to go further

AI is only as good as the instructions you give it. Here are 20 copy-paste prompts, tested and optimized for a real estate agent’s day: listings, showing reports, seller appointments, objections, social media. If you want to lay the foundations first, we’ve published a complete guide on generative AI applied to real estate.

Before you start: the 3 rules of a good prompt#

An effective prompt rests on three simple elements you’ll find in each model below.

1. The context#

Who you are, what your market is, what the situation is. Without context, AI answers as if speaking to anyone. With context, it speaks to an agent in your city, on your property type.

2. The precise task#

What you expect as a result, in what format, what length. “Write a listing” is vague. “Write a 150-word listing structured as hook / spaces / neighborhood, warm and professional tone” produces a usable result.

3. The constraints#

The tone you want, the information not to forget, what to avoid. Negative constraints (“no invention”, “no sentences over 25 words”) prevent the most frequent slips.

A vague prompt like “write a listing for a 2-bed” will produce a generic result. A detailed prompt with neighborhood, target, tone, and the property’s strengths will produce a usable result immediately. The difference takes thirty seconds more writing. The 20 prompts below work equally well on ChatGPT, Claude, and Gemini.

Listing writing, 4 prompts#

The listing is still the first touchpoint between your property and a potential buyer. The four prompts below cover creation, multi-version testing, correction, and the SEO version. For the complete method, we’ve detailed how to write punchy listings with ChatGPT, prompt by prompt.

Woman with headphones and glasses writing in a notebook in front of a laptop, illustrating ChatGPT-assisted real estate listing writing

Photo: Vlada Karpovich · Pexels

Prompt 1: the complete, structured listing#

You are an expert real estate copywriter. Write a listing for [property type, square footage] located in [address or neighborhood]. Target: [buyer profile]. Tone: professional and warm. Highlight: [3 property strengths]. Structure: hook in the first sentence, then space description, then neighborhood strengths. 150 words max. No spelling errors, no invented information.

This prompt works for all property types. The key is specifying the buyer target and specific strengths—otherwise the result stays generic.

Prompt 2: 5 variants for A/B testing#

Here’s the info on a property: [paste info]. Write 5 versions of the listing with different angles. Version 1: premium, luxury tone. Version 2: warm and family-friendly tone. Version 3: factual, data-driven tone. Version 4: neighborhood and lifestyle focus. Version 5: potential and projection focus. 100 words each.

Publish the version best suited to the target, or test two versions on two different portals to see what performs best.

Prompt 3: the corrected and improved listing#

Here’s an existing real estate listing: [paste the listing]. Fix spelling errors, improve structure, check nothing is missing. Add a hook in the first sentence. Keep the same tone but make the text more fluid and professional. Flag any important info that seems missing.

Prompt 4: the SEO description for the website#

Write an SEO-optimized description for this property: [property info]. Naturally integrate the following keywords: [neighborhood + property type + city]. The description should make readers want to click, be informative, and run 200-300 words. Add a title tag of 60 characters max and a meta description of 155 characters max.

Showing reports, 3 prompts#

The showing report is the most time-consuming and least rewarding task in an agent’s day. It’s also the first to automate: a voice memo after the showing, and AI produces a structured document, ready to send.

Top view of a laptop, coffee mug, and notebook, illustrating automatic formatting of a real estate showing report

Photo: Cup of Couple · Pexels

Prompt 5: the structured report from a voice memo#

Here’s the transcription of a voice memo after a property showing: [paste transcription]. Reformat it into a structured report with the following sections: Buyer info / Property toured / Overall impression / Positives noted / Concerns or objections / Recommended next step. Professional tone, short sentences. End with an action recommendation for the seller.

Prompt 6: the weekly seller update#

Here are the reports from the [number] showings this week on the property at [address]: [paste reports]. Write a seller summary that recaps: number of showings, buyer profiles, recurring positives, recurring objections, and recommendation for next steps (hold price, adjust, highlight certain strengths). Reassuring but honest tone.

This prompt replaces the Friday email you never have time to write. And that email is precisely what maintains seller trust over the listing’s lifetime.

Prompt 7: the ready-to-send showing report email#

Write a showing report email to send to the seller. Property: [address]. Buyer: [quick profile]. Impression: [positive/mixed/negative + details]. Next step: [what’s planned]. Email should be professional, warm, and no more than 100 words. Subject line included.

Seller appointment prep, 3 prompts#

The CMA appointment is where you win or lose the listing. Three prompts to walk in prepared: neighborhood pitch, presentation deck, and responses to the most frequent objections. For the complete methodology, here’s how to prep your next appointment in under 30 minutes.

Two people reviewing a presentation deck outside, illustrating prep for a CMA appointment with an AI-generated seller pitch

Photo: Zen Chung · Pexels

Prompt 8: the neighborhood pitch#

I’m going to a CMA appointment at [full address]. Give me: 5 sales arguments for this neighborhood (transit, shops, schools, parks, dynamics), price-per-square-foot trends over the last 12 months if you know them, the typical buyer profile searching in this area, and 3 likely seller objections with tailored responses.

Prompt 9: the presentation deck#

Build me a CMA presentation outline for this property: [property info]. The deck should contain: brokerage intro (2 lines), local market analysis, valuation methodology, our commitments (visibility, follow-up, visual quality), and next steps after signing the listing agreement. Professional, confident tone.

Prompt 10: seller objection responses#

I’m a real estate agent. Here are the 5 objections I hear most often in seller appointments: 1) “I’d rather wait for spring to sell” 2) “The neighbor sold for more” 3) “I don’t want an exclusive listing” 4) “Your commission is too high” 5) “I’ll try to sell on my own first.” For each, write a diplomatic, factual, convincing response. Use market data when relevant.

A well-built prompt is thirty seconds more writing, and thirty minutes less on the task.

Buyer objection handling, 2 prompts#

On the buyer side, two situations recur: hesitation after a showing, and price negotiation. Two prompts to keep the conversation alive and defend the price without pushing back too hard.

Prompt 11: responding to a hesitant buyer#

A buyer toured [property type] at [address] and is hesitating. Their concerns: [list concerns]. Write a follow-up email that addresses each concern, provides concrete responses, and proposes a second showing or phone call. Tone: empathetic but proactive. 120 words max.

Prompt 12: price negotiation#

A buyer offers [offered price] for a property listed at [list price]. The gap is [X%]. Write an argument to justify the price to the buyer, based on: neighborhood comps, the property’s specific strengths, and the local market dynamic. Firm tone but open to discussion.

Social media content, 3 prompts#

Instagram, LinkedIn, and TikTok have become major acquisition channels for agents who get serious about them. AI doesn’t write the strategy for you, but it kills the blank page and saves you hours on production.

Presentation in front of a screen titled Digital Evolution with mentions of Social Media, Artificial Intelligence and Big Data, illustrating social media content for real estate agents

Photo: RDNE Stock project · Pexels

Prompt 13: the monthly content plan#

You are a real estate social media expert. Build me a calendar of 8 Instagram posts for next month. My brokerage specializes in [market]. My target: [profile]. 4 educational Reels (AI tips, market breakdown) and 4 photo posts (property for sale, brokerage behind-the-scenes, client review, neighborhood). For each post: the hook (first line), the angle, and the CTA.

Prompt 14: the Reel hook#

Give me 10 Instagram Reel hooks for a real estate agent talking about [topic]. Hooks should be short (under 10 words), negative or provocative (stops the scroll), and make people want to watch more. No “did you know that”, no soft rhetorical questions.

Prompt 15: the Instagram caption#

Write an Instagram caption to go with [a property photo / a Reel / a tip]. Tone is professional but accessible, casual you. Structure: hook in the first line (the only line visible before “see more”), 3-4 lines of content, open question as CTA. Add 5 relevant real estate hashtags. 150 words max.

Document analysis, 3 prompts#

HOA minutes, energy disclosures, purchase agreements: three documents an agent has to read fast, and well. With these prompts, you go from a full hour’s read to a usable summary in two minutes.

Prompt 16: HOA minutes summary#

Here are HOA meeting minutes [import PDF]. Give me: the assessments voted and their amounts, the impact on dues, ongoing disputes or tension points, and the info a potential buyer absolutely needs to know before buying in this HOA.

Prompt 17: energy disclosure analysis#

Here’s an energy disclosure [import PDF]. Summarize in 5 key points: the energy rating, main consumption sources, recommended improvements and estimated cost, impact on property value, and the arguments I can use in a seller appointment to explain this disclosure.

Prompt 18: purchase agreement review#

Here’s a purchase agreement [import PDF]. Identify: contingencies and their deadlines, unusual or potentially problematic clauses, attention points for buyer and for seller, key dates to remember. This doesn’t replace a real estate attorney’s review, but lets me walk in prepared.

Email & prospecting, 2 prompts#

The personalized follow-up and the neighborhood newsletter are two levers under-used by most brokerages. Done well, they’re also two of the most profitable channels to prospect with AI without saturating your contacts.

Prompt 19: the personalized follow-up#

Write a follow-up email for [a buyer who toured 5 days ago / a seller who didn’t follow up on a CMA / a cold lead who didn’t respond]. Context: [what happened]. The email should be short (80 words max), personalized (mention a specific element from our exchange), and propose a clear next step. No canned phrasing, natural tone.

Prompt 20: the neighborhood newsletter#

Write a monthly newsletter for my contact list on the [neighborhood/city] market. Content: 1 standout local market fact this month, 1 practical tip for owners, 1 favorite listing from our portfolio (I’ll fill in), 1 open question to spark conversation. Warm, expert tone, casual you. 250 words max.

How to go further#

These 20 prompts are a starting point. To get the most out of them, two recommendations.

First, personalize them with your brokerage’s specifics, your market, and your communication tone. A generic prompt will always produce a worse result than one adapted to your context. Your best listings, best showing reports, best emails are already the raw material—just paste them into a prompt to give AI your style.

Second, don’t repeat context every time. You can configure your Claude Projects with your brokerage profile, templates, markets, and examples: AI will already know your frame, and your prompts will become much shorter and more precise.

The productivity leap doesn’t come from the isolated prompt—it comes from mechanizing prompts into your daily work. The test: if you’re typing more than 30 seconds to explain context to AI, it’s time to turn it into a Claude Project.

Questions we get asked.

How much time does an AI prompt save?

A well-built prompt saves 30 minutes per task on average (writing a listing, showing report, email). Over an agent’s week, that’s 6-10 hours recovered.

Which real estate agent tasks can be automated with ChatGPT?

Writing listings, showing reports, follow-up emails, seller appointment prep, HOA minutes analysis, social media content. Roughly 37% of tasks per Morgan Stanley.

ChatGPT or Claude for real estate prompts?

Both work. ChatGPT is better at creative writing and listings (natural, varied tone). Claude excels at analyzing long documents (100-page HOA minutes, purchase contracts). In practice, agents who succeed use ChatGPT for daily production and Claude for technical analysis. Budget $20/month each for the Plus/Pro versions, so $40/month for both.

Have something to say?No comments yet

Share your take, ask a question, or tell us about your brokerage experience. Moderation is fast.

Your email stays private. No spam.
Thanks for your comment. It will appear after moderation.
Be the first to comment. The best discussions come from precise questions and field experience.
· 2026 · Article 09 · Prompts & Use Cases Published July 2026
See Kappn in 30 min

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *