A chatbot for real estate answers the questions buyers and renters send through your website and listing pages, captures their details and books inspections or calls into your calendar, including at night when nobody on the team is free. It works well for factual questions and scheduling, and it should hand over to an agent for anything that needs judgement, negotiation or advice, and for anything that touches who a buyer is rather than what they want.
What a real estate chatbot is good at
Most listing enquiries are short and repetitive. Is it still available? Can I see it this weekend? Does it allow pets? What are the strata or association fees? An agent who is at an open home, driving or asleep cannot answer these in minutes, and the person asking has usually sent the same message to two or three other agents.
A chatbot suits the repetitive part of that work:
- Confirming the facts you have already published for a listing: address, price guide, bedrooms, inspection times, how to apply.
- Asking the two or three questions an agent would ask first, such as whether they are buying or renting, their timeframe and the best number to reach them on.
- Offering open inspection or private viewing times from your real calendar and confirming the booking in the chat.
- Writing the conversation and contact details into your CRM so the agent sees the full history before calling back.
It is poor at anything that is a matter of opinion or negotiation: what a property will sell for, whether an offer is good, or whether a suburb is “safe”. Those belong to an agent.
What it should not answer
Write this list down before the chatbot goes live. A good rule is that the bot answers from approved listing information and your own published policies, and says “I will get an agent to confirm that” for everything else.
- Price negotiation and offers. It can share the advertised price. It should not hint at what a seller will accept.
- Neighbourhood descriptions based on who lives there. This is the area that creates fair housing risk, covered below.
- Legal, tax, mortgage or financing advice. It can point to a licensed professional.
- Anything about a person’s circumstances such as family status, income or background. It should not ask about these unless a law or lender requirement makes it necessary, and then a person should handle it.
Fair housing rules for a chatbot in the United States
In the United States the Fair Housing Act prohibits discrimination in housing based on race, colour, national origin, religion, sex, familial status and disability, according to the US Department of Housing and Urban Development (HUD) overview of the Act. The Act’s advertising provision, found in 42 U.S.C. 3604(c), makes it unlawful to publish any notice, statement or advertisement about a dwelling that indicates a preference or limitation based on those categories. The full text of section 3604 is published by Cornell Law School.
A chatbot speaks on your behalf, so what it says is a statement about the dwelling. In practice that means:
- Do not let it describe a home or suburb as suited to, or not suited to, a type of family, religion, nationality or age group.
- Do not let it steer. If a buyer asks “is this a good area for people like us”, the bot should give factual information such as schools’ official pages, commute times and amenities, and offer an agent call, without characterising residents.
- Give every visitor the same answer to the same question, and test this by asking it the same thing in different ways.
- Keep a log of conversations so you can review what was said.
This is a general summary and not legal advice. State and local laws and your brokerage’s policies may add further rules, so check them with your broker or a lawyer before you launch.
Capturing the lead and booking inspections
A chatbot earns its place when a conversation ends with something the agent can act on. The sequence that works for most agencies is short.
- Answer the first question straight away, from the listing.
- Ask whether they would like to see the property, and offer the next available times.
- Ask for a name and mobile number to confirm the booking, at that point and not before. People hand over contact details when they get something useful in return.
- Confirm the booking by text or email and add it to the agent’s calendar.
HighLevel’s help centre describes how its Conversation AI bot books appointments: it offers open slots from a connected booking calendar, confirms the booking in the conversation and checks for conflicts with the contact’s existing appointments, and it needs a booking calendar and a trained bot before it works (HighLevel, appointment booking in Conversation AI, as at November 2026). That page does not describe a general handover to a person, so if you use that tool, confirm how a live agent takes over before you rely on it. CentriWeb builds many client systems on GoHighLevel, so weigh that in how you read this example. The same steps apply to other chat tools.
Hand-off to an agent
The hand-off is the part buyers notice. Set clear triggers for passing the conversation to a person:
- The visitor asks for a person, or the bot has failed to answer twice in a row.
- The question is about price negotiation, an offer, finance, a legal condition or a complaint.
- The visitor sounds frustrated or the matter is urgent, such as a tenant reporting a problem with a rental.
When it hands over, the agent should receive the whole conversation, the listing the person was asking about and the contact details, so the buyer never repeats themselves. During business hours the bot can alert the agent by text. After hours it should tell the visitor when to expect a reply, and that time should be one the team can keep.
Connecting the chatbot to your CRM
A chatbot that keeps its leads inside its own dashboard creates a second list for the team to check. Connect it so every conversation creates or updates a contact in the CRM your agents already use, tagged with the listing and the source. If your CRM is already set up, ask whether the chatbot can write to it directly or through an automation tool.
Check three things before you go live: that duplicate contacts are merged, that the listing address travels with the lead, and that the agent assigned to the listing is the one notified. Our guide on real estate lead follow-up scripts covers what the agent should say in the first call back.
A worked example
Suppose an agent with fifteen active listings gets around forty enquiries a week through the website and listing portals, and a quarter of them arrive between 7pm and 7am. That is about ten enquiries waiting overnight, each one competing with other agents’ replies. These numbers are an example only: use your own enquiry log to count.
With a chatbot, each of those ten gets an instant answer and an offer of inspection times. Say six book a time. The agent starts the day with six confirmed inspections and four conversations to continue, instead of ten cold enquiries to chase. The value does not come from the bot being clever. It comes from nobody waiting overnight for an answer to a factual question.
Who a chatbot does not suit
A chatbot may not be worth setting up if you receive a handful of enquiries a week and answer them all within minutes, or if most of your business comes from referrals and phone calls. In that case a good contact form and a reliable text reply may be enough. A chatbot also needs someone to keep listing information current: a bot quoting last month’s price does more harm than no bot.
If you want a system built around your listings, calendar and CRM, and looked after afterwards, CentriWeb does this as part of its AI answering and lead follow-up service. For the wider picture on chat tools, see our guide to an AI chatbot for a small business website, and for what AI can do across an agency, AI for real estate agents. Property managers have a related set of needs on the property industry page.