ChatGPT prompts for real estate agents: 16 copy-paste prompts built from what real agents do (2026)

What are the best ChatGPT prompts for real estate listing descriptions?

Five agents here use it for listings, in two different ways. JJ Johannes and Amy Vastardis start from ChatGPT's draft and edit it; Krista Bradley writes her own and only has ChatGPT polish it and suggest angles she missed. Joe Vairo uploaded the transcript of every video he has posted so drafts sound like him, and Brandy Johnson uses it for fresher wording and a longer bio for listing presentations. For more on this one job, see Can ChatGPT write real estate listing descriptions? 9 real agents.

Write a listing description for a home in {CITY}. Facts: {BEDS, BATHS, SQUARE FEET, LOT, STREET, UPDATES, FEATURES}. Write 120 to 180 words, warm but plain, opening with the strongest feature. Use only the facts given; never invent features, room sizes, prices, school names or neighborhood claims. If something a buyer would expect is missing, write [MISSING]. Don't describe who the home is ideal for in terms of family status or other protected traits.

Write a first draft of listing remarks for {ADDRESS OR AREA}. Facts: {BEDS, BATHS, SQ FT, UPGRADES, LOT, NEIGHBORHOOD FEATURES}. Character limit: {LIMIT}. Match the tone of these remarks I wrote: {1-2 SAMPLES}. Use only the facts I gave; never invent features, measurements, school names, prices or distances. Mark anything needed but not given as [MISSING]. Describe the home, not who the buyer should be.

Here is my listing description, written in my own voice: {PASTE DRAFT}. Polish it lightly: fix flow and grammar, keep my wording and tone, don't add any facts and don't use "nestled", "exceptional" or "ideal". Then, separately, list 3 angles or features of the home or neighbourhood I didn't mention that a buyer might care about (as questions for me to check, not claims). Finally, write 3 opening lines for a 30-second listing video, each under 15 words.

Use the transcripts and listings in this project as my voice and my ideal client profile. New listing facts: {ADDRESS, BEDS/BATHS, SQ FT, FEATURES, NEIGHBORHOOD}. Photo file names with a few words each: {PHOTO LIST}. Write: 1) an MLS description under {CHARACTER LIMIT} that works in the search phrases buyers use for {PROPERTY TYPE} in {CITY}, 2) the 10 photos to lead with, in order, and why, 3) a 45-second voiceover script for the listing video. Use only the facts given; never invent features, schools, distances or measurements, and follow fair-housing rules.

Part 1 (listing): Here is my draft listing description and the MLS facts for {ADDRESS}: {DRAFT} / {FACTS}. Rewrite it in under {WORD LIMIT} words with fresher wording than 'stunning', 'nestled' or 'must-see'. Use only the facts given; mark anything you'd want to add as [ASK AGENT]. No words about who the home suits (families, couples, age) or about schools' quality. Part 2 (call script): Write a friendly call script for my assistant, {NAME}, calling {LEAD TYPE} who {HOW THEY CAME IN}. Goal: {GOAL}. Include a 2-sentence opener, 3 questions, answers to the 3 most likely objections and a closing line that books a call with me. Keep it conversational; no pressure tactics.

What ChatGPT prompts help real estate agents with pricing, market numbers and a business plan?

Upload the reports you already pull. Chuck Bachteler downloads property, neighborhood and market reports from RPR, runs seven prompts in order to reach a price, and turns the output into a seller presentation in Gamma. Frank Trelles asks for mortgage payments and returns on the spot and has it show its work; per CNN the tool has struggled with some basic math, so he treats the answer as an approximation. Carlos Melendez uploaded his transaction roster, let ChatGPT ask him the questions it needed, and got a draft business plan.

Using these RPR reports for {ADDRESS} ({BEDS} bd / {BATHS} ba / {SQFT} sq ft), summarise days on market, price per square foot and recent sales for comparable homes within {RADIUS} in the last {MONTHS} months. Then suggest a list-price range and explain each assumption, and give a 6-slide outline (headline + 3 bullets each) I can paste into Gamma for the seller presentation.

Estimate the monthly mortgage payment for a {PRICE} purchase with {DOWN PAYMENT} down, at {RATE}% interest, amortized over {YEARS} years, with {POINTS} origination points at closing. Show the formula and each step so I can check it. Then estimate a simple annual return on investment using rent of {RENT} and expenses of {EXPENSES}. Use only the numbers I give; never assume taxes, insurance or fees. If a number is needed but missing, write [MISSING]. Label every result as an estimate.

This file is my transaction roster as a real estate agent in {MARKET} for {YEARS}. First, tell me which areas and price ranges I've done the most business in, using only the file. Then ask me the most important questions you need answered to draft my {YEAR} business plan, and wait for my answers. After I answer, draft the plan. Never invent figures, market stats or results. Label any growth number as a goal, and mark anything you need from me [ASK].

What ChatGPT prompts work for real estate lead generation and follow-up?

Both stories here are about answering fast with what the person already told you. Adrian Dukes connected ChatGPT to his Facebook page as a chatbot that asks what visitors are looking for, alerts his phone and can set up the appointment or send a Google Form or MLS link. Missy Plunkett has ChatGPT draft personalized follow-ups from past conversations with each client. Texas REALTOR magazine reminds Texas agents that TREC rules still require the Information About Brokerage Services form at the first substantive conversation about a property.

You help me, {AGENT NAME}, a real estate agent in {AREA}, reply to Facebook messages. Here is the conversation so far: {PASTE MESSAGES}. Draft a short, friendly reply that answers only what they asked using these facts: {FACTS}, asks the one missing detail I need most (buying or selling, area, price range or timeline), and offers two appointment times: {TIMES}. If they ask about a specific property, include this MLS search link: {LINK}. Don't guess prices, schools or anything not in my facts; write [CHECK] instead. Under 80 words.

Here are my past messages with this buyer: {THREAD}. Draft a short follow-up after today's showing that mentions what they cared about and suggests next steps.

What ChatGPT prompts help with client emails, contracts and hard conversations?

Andres Asion tells ChatGPT he is the buyer's agent and asks it for any red flags in a contract that might hurt his buyer, then reads it all himself; he also had it rewrite a client's ignored complaint email around liability, and has lawyers approve the addendums it helps draft. Chris Saxon keeps a ChatGPT project for each client and asks for a summary, the decisions to make and an agenda before a call. Kasey Jorgenson's brokerage uses ChatGPT for emails and newsletters and says AI doesn't replace an attorney on agreements.

I am the {BUYER'S OR SELLER'S} agent on this contract and addendum for {PROPERTY TYPE}. List any sections that might hurt my client: quote the exact clause, give the section or page number, and explain in one plain sentence why it needs attention. Don't summarize the whole contract. Never invent clauses, deadlines, dollar amounts or laws; if something seems missing, write [MISSING]. If a point needs a lawyer's opinion, write [ASK: lawyer]. This is a review aid, not legal advice.

Rewrite this email from {CLIENT ROLE} to {RECIPIENT} about {PROBLEM}. Original email: {EMAIL}. Timeline of contact attempts: {TIMELINE}. Make it firm, factual and polite, under 250 words, and point out the liability concerns of leaving the problem unresolved. Ask for a reply by {DATE}. Use only the facts given; never invent dates, laws, statutes, contract terms or damages. If a fact is missing, write [MISSING]. If a legal claim would need a lawyer to confirm, write [ASK: lawyer].

You're helping me prepare for a call with my client {CLIENT FIRST NAME} about {ISSUE, e.g. a septic problem found in the pre-list inspection}. Using only the documents and notes in this project and what I just told you, write: 1) a short plain-English summary of where things stand, 2) the decisions the client needs to make, with the options and what each costs according to the estimates, and 3) a 3-5 item agenda for our call. Don't invent costs, dates or contractor opinions; if something is unclear, list it as a question for the call.

You help agents at {BROKERAGE} in {MARKET} write {JOB: a client email / the monthly newsletter / an internal policy}. Our tone: {TONE}. Facts to use: {FACTS}. Write a draft under {LENGTH} words using only these facts. Do not invent numbers, laws, dates or listing details; if something is missing, write [CHECK]. End with a short list of what a person should verify before this goes out.

What ChatGPT prompts help real estate agents with blog content and getting better at sales?

Carrie Little has ChatGPT turn her own YouTube videos into blog posts, and checks each one against license law and the Code of Ethics. Chris Saxon trained ChatGPT on the buyer process he follows and sends it the transcript of a consultation (recorded with the Granola note-taker) to grade how he did on each of the 16 steps.

Here is the transcript of my YouTube video for {BROKERAGE NAME} in {TOWN}: {TRANSCRIPT}. Turn it into a blog post of about 600 words for buyers and sellers in {TOWN}: a clear title, three short sections with subheadings, and a closing line inviting readers to watch the video and contact me. Keep my wording and examples; use only facts that are in the transcript, and mark anything you're unsure of as [CHECK]. No claims about schools, safety or who a neighborhood is 'perfect for'.

Here is my buyer consultation process, step by step: {YOUR NUMBERED STEPS}. Here is the transcript of today's consultation: {TRANSCRIPT}. For each step, grade me A-F, quote the part of the transcript that shows it (or say the step was skipped), and give one specific thing to do better next time. Don't invent anything the client said. End with the 3 biggest improvements for my next consultation.

What should real estate agents watch out for with ChatGPT?

• Made-up facts. Bachteler says AI "can lie with confidence." Check every measurement, feature, school, fee and number against the MLS and your documents before it's published or sent.
• Fair housing. Describe the home, not who it's "ideal for"; the prompts above tell ChatGPT to avoid family-status and other protected-trait language, but you're responsible for the final copy.
• Contracts and legal questions. Asion uses ChatGPT to point to sections and sends drafted addendums to lawyers; treat its answer as a review aid, not legal advice.
• Math. Trelles uses it for approximations, not exact numbers; have your lender confirm any payment you quote a client.
• Client data. As a suggestion, check your brokerage's AI policy and your ChatGPT plan's data settings before uploading contracts, transaction rosters or client threads.