Skip to main content

AI for Non-Technical Real Estate Agents: Where to Start

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. 4 years coaching at Tom Ferry, 9 years prior as a client.

AI for non-technical real estate agents starts with a skill you already have: explaining a task clearly, like briefing a new assistant. You don’t need code, integrations, or ten tools. One chat app, your voice, and one weekly task are enough. This guide gives the five-step system, the first prompt, and the review step that protects your license.

Key Takeaways

  • Using AI well is a briefing skill, not a technical skill, and agents brief people all day.
  • Start with one task you already do every week, not with a list of tools.
  • Talk to the AI instead of typing, because voice removes the biggest barrier for non-technical agents.
  • Save the brief that works so you never explain yourself twice.
  • Nothing AI drafts goes public until you’ve checked the facts, the fair housing language, and your broker’s name.

What does “non-technical” actually mean with AI?

Using AI as a non-technical agent means working in a plain chat app like ChatGPT, Claude, or Gemini. You type or speak instructions in everyday language. There’s no code, no software integration, and no automation to build.

If you can text a client and brief a transaction coordinator, you have every skill the first 90 days require. The technical layer comes later, if it comes at all. That layer means connecting AI to your CRM and building automated workflows.

Why this matters for real estate agents

Most agents are stuck at one of two points, and neither one is a technology problem. According to NAR’s 2025 Technology Survey (September 2025), 32% of agents hadn’t used AI in their business at all. Only 17% reported a significantly positive impact, and 46% said AI had no noticeable impact.

Read those together. A third of the industry hasn’t started. Nearly half of the rest are using AI and getting nothing measurable from it. Both groups share one cause: they treated AI as a tech project instead of a delegation problem.

The same NAR release frames technology around what it protects: in NAR Deputy Chief Economist Jessica Lautz’s words, “the trusted relationship between the agent and client.” That’s the point of all of this. AI should buy back the hours you spend drafting so you can spend them with clients.

“Every agent who tells me they’re not technical has briefed a transaction coordinator, a stager, and a nervous first-time buyer in the same afternoon. That’s the whole skill. AI needs the same brief, written down once.”
— Emily Terrell, Tom Ferry Coach

How do non-technical agents start using AI? The 5-step system

Which task should you hand to AI first?

Pick one task you do every week that’s mostly writing and that you’d happily delegate. Strong first candidates include:

  • first drafts of listing descriptions
  • follow-up texts to leads who went quiet
  • recap emails after showings
  • weekly social captions
  • a monthly market note to your sphere

Use three filters. The task repeats weekly. You already know what a good version looks like. And it costs you at least 30 minutes a week. Don’t start with your hardest problem. Start with the one you’ll repeat often enough to get good at.

Should you type or talk to AI?

Talk. The fastest way past “I’m not technical” is to stop typing. ChatGPT Voice lets you speak your request and hear the answer in the mobile app or on the web. Talk it through the way you’d leave a voicemail for your assistant: the property, the client, the goal, and what you’d say if you had time to write it yourself.

OpenAI’s own documentation notes that ChatGPT can make mistakes and that important information should be checked. That’s why Step 5 exists. Set your voice up before you drive, never while the car is moving.

What do you actually say to AI?

Give it the same four-part brief you’d give a new assistant:

  • Who you are: your market and your role.
  • What you need: the task and its length.
  • Who it’s for: the reader.
  • What good looks like: paste an example you’ve written that you like.

The example matters most, because it teaches your voice faster than any description of your voice.

Here’s a first prompt you can copy:

I’m a real estate agent in San Antonio. Write a first draft of a listing description for the home below. The readers are move-up buyers comparing homes in this price range. Keep it under 150 words, use no exclamation points, and don’t describe who the home is “perfect for.” Match the tone of this description I wrote: [paste your best description]. Use only these property facts: [paste your MLS input].

Then keep the conversation going. “Shorter.” “Lead with the kitchen.” “Less formal.” Revising is part of using AI, not a sign it failed.

How do you stop re-explaining yourself every time?

Save the brief once it works. In Claude, a Project holds your uploaded documents and your instructions and applies them to every chat inside it. Projects are available on free accounts, which can create up to five. If you work in ChatGPT, use its saved-workspace equivalent. The principle is identical.

What goes into a saved setup:

  • three samples of your best writing
  • your market and specialties
  • your brokerage name exactly as it must appear
  • a short list of words you never use

That setup is the difference between re-explaining yourself for 20 minutes every time and getting a usable draft in one message.

What do you check before anything goes out?

Check three things on every draft before it reaches a client or a feed.

  1. Property facts. Check every fact against the MLS input, not against the draft. AI will describe a feature that doesn’t exist with complete confidence.
  2. Fair housing language. Remove phrasing that signals a preference about who should live in the home. NAR’s Fair Housing resources are the reference.
  3. Broker identification. In Texas, TREC Rule 535.155 treats social posts and emails as advertising, and AI drafts almost never include your broker’s name.

For the state-by-state detail, read what actually risks your license with AI-generated content. This is general information, not legal advice. Confirm specifics with your broker or attorney.

How I use this in my own business

I close 70+ transactions a year in San Antonio in about five hours a week of active management. I’m not an engineer. I’m an agent with ADHD who built systems because I didn’t want to babysit my business.

The first AI workflow that stuck for me wasn’t clever. It was talking. After a showing, I open voice mode before I pull out of the driveway. I talk through what the buyers loved, what killed it for them, and the questions they asked. By the time I reach the car line at school, I have a follow-up text drafted and a note ready for my CRM.

That’s still how I start agents who tell me they’re not tech people: one task, spoken, saved, and reviewed. When I teach this live to a brokerage, the room builds it on its own listings, not on a demo property. The full session design is what AI training for real estate agents must cover.

Common mistakes

  1. Starting with a tool list instead of a task. Five tools with no workflow produce nothing. One task with one tool produces a habit.
  2. Asking one-line questions and judging AI by the answer. “Write a listing description” gets a generic copy. The four-part brief gets your copy.
  3. Re-explaining yourself in every chat. If you’re typing your market and tone for the fifth time, save the brief.
  4. Pasting sensitive client data into a consumer chat app. Financial details and personal identifiers stay out of general-purpose tools.
  5. Publishing straight from the chat window. Every draft goes through the three-point review, every time.

Frequently Asked Questions

Do I need to be good with technology to use AI as a real estate agent?

No. Using AI as a real estate agent requires clear instructions, not technical skill. If you can text a client, brief a transaction coordinator, and describe a listing out loud, you can get useful output from ChatGPT or Claude. The technical layer, meaning CRM integrations and automations, is optional, and you’ll get real value long before you touch it.

What’s the easiest AI tool for a real estate agent to start with?

The easiest AI tool is a general chat app you can talk to. ChatGPT is the most common starting point, and NAR’s 2025 Technology Survey ranked it first among the AI tools agents use, at 58%. Claude is a strong choice for writing in your own voice. Pick one, use it on one weekly task for 30 days, and skip tool comparisons until then.

Can I use AI on my phone?

Yes, and for non-technical agents the phone is often the better place to start. ChatGPT’s Voice feature lets you speak your request and hear the answer in the mobile app or on the web. Talk through a showing, a listing, or a follow-up between appointments, then refine the draft at your desk. Set it up before you drive, not while you’re driving.

How long does it take to learn AI as a real estate agent?

The first useful draft takes one conversation. Consistent results take a few weeks of using the same brief on the same task and correcting the output each time. That repetition is where the learning happens: you’re refining your instructions, not learning software. Agents who switch to a new tool every week tend to stay at the beginning indefinitely.

What should I never put into an AI tool?

Never paste a client’s financial details, Social Security number, account information, or other sensitive personal data into a consumer AI chat app. Many platforms retain what you enter, so check a tool’s data settings before using it for client work. Property facts, your own writing samples, and anonymized situations like “a buyer relocating for work” are safe starting inputs.

Do I have to tell clients I used AI to write my marketing?

In most states, there’s no general requirement to disclose that AI drafted your listing copy or captions. What’s required is that the content is accurate, not misleading, and compliant with fair housing and advertising rules like TREC Rule 535.155 in Texas. Rules vary by state and change often, so confirm with your broker. This is general information, not legal advice.

Will AI replace real estate agents?

No. AI drafts; the agent decides, negotiates, and carries the license. What’s changing is the gap between agents who use AI to get hours back and agents who don’t. Those hours go into work AI can’t do: pricing conversations, negotiations, and the client relationships that produce referrals. AI isn’t replacing you. The agent who knows how to use it is.

Bring this to your team or event

Emily Terrell speaks at brokerage events, real estate conferences, and team trainings on AI, systems, and social media — the exact playbook in this post, delivered live to your audience. As a Top Coach and Speaker at Tom Ferry International and an active agent closing 70+ transactions a year, Emily speaks from the stage about what’s working right now, not theory. Recent stages include NAHREP and eXp Con. See her keynote topics.

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

For real estate agents who want to implement this: Get the weekly real estate prompt library at weeklyrealestateprompts.com or follow @coachemilyterrell on Instagram for daily systems and AI breakdowns.