What should a real estate agency automate with AI first?
Start with high-volume, low-risk work: responding to and qualifying new sales and rental enquiries, booking inspections and keeping the CRM up to date. Next, add appraisal follow-up and maintenance request intake with clear escalation rules. Leave vendor reports until last and run them draft-first. Pricing advice, negotiation, tenant selection and urgent repair decisions stay with people.
Key takeaways
- Rank workflows by weekly volume, how repetitive they are and what happens if the AI gets it wrong.
- New enquiries are usually the best first workflow: they are frequent, time-sensitive and follow a script.
- Maintenance intake and vendor reporting work well as capture-and-draft tasks, with a person deciding and sending.
- Out-of-hours enquiries from interstate and overseas buyers need time-zone-aware replies and callbacks.
- Tenant and buyer details are personal information: check your Privacy Act position before connecting AI tools.
Which agency workflows can AI handle?
In a real estate agency, an AI employee is best thought of as a team member with a narrow job description: answer or call enquiries, ask set questions, look up approved information, write notes into the CRM, draft messages for approval, and report back. Across a typical agency, five workflows come up again and again:
- Sales enquiries: portal and website enquiries that need a fast reply, a few qualifying questions and an inspection booking.
- Appraisal follow-up: owners who had an appraisal but have not listed, and need timely, relevant contact.
- Rental enquiries: questions about listings, inspection times and how to apply.
- Maintenance requests: tenants reporting problems that need to be captured, logged and routed.
- Vendor reporting: weekly updates on enquiries, inspection numbers and buyer feedback.
How should you prioritise them?
| Workflow | What the AI handles | What stays with people | Risk if it goes wrong | Suggested order |
|---|---|---|---|---|
| Sales enquiries | First reply or call, qualifying questions, inspection booking, CRM update, lead summary | Pricing advice, negotiation, offers | Low to medium | First |
| Rental enquiries | Answers from approved listing details, inspection booking, application link | Assessing applications, tenant selection | Low to medium | First, if you run a rent roll |
| Appraisal follow-up | Reminders, drafted check-ins, logging responses | Price opinions, the listing conversation | Medium: marketing consent rules apply | Second |
| Maintenance intake | Capturing the problem, address and access details, logging a task, alerting the property manager | Deciding urgency, instructing tradespeople, landlord approvals | Medium to high for urgent issues | Second, with escalation rules |
| Vendor reporting | Pulling enquiry and inspection data from the CRM into a draft report | Interpreting feedback, advice, sending | Medium: accuracy matters | Third, draft-first |
How do you pick the first workflow?
- List the workflows above that apply to your agency and estimate weekly volume for each.
- Note where the data lives: portal emails, website forms, the CRM, a property management system, a shared inbox.
- Score each for risk: what is the worst realistic outcome if the AI gets one wrong?
- Choose the highest-volume, lowest-risk workflow whose data is already in a connected system.
- Define one measure of success before you start, such as time to first response or inspections booked.
- Run a pilot, review every conversation for the first weeks, then expand.
Why are sales enquiries usually first?
New enquiries are frequent, time-sensitive and follow a predictable script, and every one should end with an updated CRM record. An AI voice agent can call each new enquiry, ask about budget, timeframe and finance status, and book an inspection. The step-by-step setup, script table and call rules are in our deep dive on automating real estate qualifying calls.
How should appraisal follow-up be automated?
Appraisal follow-up is a mix of reminders and relationship. The AI can create follow-up tasks, draft a check-in when a comparable property sells, and log replies, while the agent sends or approves anything that discusses price. Marketing emails and texts fall under the Spam Act: the ACMA says you need consent, must identify the sender, and must honour unsubscribe requests within 5 working days. Our comparison of CRM automated follow-up tools covers sequences, built-in CRM AI and AI agents for this job.
What about rentals and maintenance requests?
Rental enquiries work like sales enquiries with different questions: move-in date, who will live there, pets, and which inspection suits. Maintenance intake is different because some requests are urgent. Treat it as capture and route: the AI records the problem, the property, access details and the tenant's contact number, logs a task, and alerts the property manager straight away when the report matches your urgent list. A person decides what happens next.
Both involve personal information. The OAIC's guidance says privacy obligations apply to any personal information input into an AI system, and to output that contains personal information. Most small businesses with annual turnover of $3 million or less are not covered by the Privacy Act, but some are, so check your position. Our Privacy Act checklist for using AI with customer data goes through it.
How do you handle interstate and overseas buyers who enquire out of hours?
Enquiries from other time zones often arrive when the office is closed. NSW is on Australian Eastern Standard Time (UTC+10), moving to daylight time (UTC+11) in summer, while Western Australia stays on UTC+8 because it does not observe daylight saving. In 2026, NSW daylight saving starts on Sunday 4 October, after which a 9am call from Sydney reaches Perth at 6am. Overseas the gap widens: the UAE is 4 hours ahead of GMT, so Dubai is 6 hours behind Sydney in standard time and 7 hours behind during daylight saving. An enquiry sent at 7pm in Dubai lands in Sydney at 2am in summer.
A sensible pattern:
- Send an immediate written reply with the listing details, inspection times and how to request documents.
- Record the buyer's location and time zone in the CRM.
- Offer a booking link or schedule the callback inside the buyer's local business hours. The telemarketing standard's calling hours are based on the time where the person being called is.
- Give the agent a summary first thing, so the human follow-up starts with context.
Where does Wharq fit?
Mia is Wharq's AI employee role for real estate agencies, and Atlas is the role for buyer's agents; both are listed on our services page. They can answer or call enquiries, ask qualifying questions, search your documents and notes before responding, draft follow-ups, update the CRM or task board and send a daily report. Sensitive or external actions can run draft-first with approvals, as described on our about page. If you would like help choosing your first workflow, email us to book a workflow audit.
General information, not legal advice. Check how the Spam Act, the telemarketing standard and the Privacy Act apply to your agency.
AI for real estate agencies: frequently asked questions
Will AI replace property managers or sales agents?
No. It takes over repetitive steps such as first responses, booking, data entry and drafting. Advice, negotiation, tenant selection and decisions about urgent repairs stay with people.
Can AI write vendor reports?
It can draft them from CRM data such as enquiry numbers, inspection attendance and recorded feedback. An agent should check the draft and send it, because vendors rely on the numbers and the interpretation.
Should AI handle urgent maintenance requests?
It should capture them and alert a person immediately, not decide what to do. Agree a list of urgent issues and an escalation path before switching it on.
Does AI work with property management software?
It depends on whether the software offers API or integration access. Check this with your vendor first, and confirm the integration during scoping.
How long does it take to set up the first workflow?
It depends on how clean your enquiry sources and CRM are and how many systems are involved. Start with one workflow and a short pilot rather than automating everything at once.
Sources
- Do Not Call Register (ACMA), "Industry Standards" (accessed October 2026)
- Australian Communications and Media Authority, "Avoid sending spam" (accessed October 2026)
- OAIC, "Guidance on privacy and the use of commercially available AI products" (published 21 October 2024, updated 17 January 2025)
- OAIC, "Small business" (accessed October 2026)
- NSW Communities and Justice, "Definitions of time and how time is regulated" (accessed October 2026)
- NSW Government, "Daylight saving" (accessed October 2026)
- UAE Government portal, "Fact sheet" (accessed October 2026)