Enquiry replies
- Each portal enquiry gets a reply drafted from the listing and its documents.
- Questions about price go to the agent, not into the draft.
- It waits for a person to check and send it.
Start with one job. Prove AI pays. Keep your clients’ data safe.
For Sydney sales agencies with 5 to 100 staff, where the week fills up with follow-ups, enquiries and vendor reports.
The Korbai Method: find the one job where AI pays in your agency, prove it on your own listings, and build from there. AI drafts behind the scenes, and a person sends anything a buyer or vendor reads.
Not now AI setting the price estimate, and AI voices calling your buyers
Start here
“It never felt like a sales pitch, it felt like a conversation with someone who knew what he was talking about.”
David Murphy Director & Residential Sales, Murphy ResidentialFixed fee, fixed window
Generally the best first step. We follow a listing from the appraisal to settlement, find where the follow-ups and reports eat the week, and put a number on the one worth handing over.
Not now AI setting the price estimate, and AI voices calling your buyers
First workflow live in weeks
We build the workflow that won the Audit, or the one you came to us for, on the CRM you already run. One thing at a time, proven on your own listings before the next.
Checked again after 30 days
Your salespeople and admin team learn to check and send what it drafts, and to keep buyer and vendor details in the tools you’ve approved.
Ongoing, with regular check-ins
Ongoing advisory. We watch the drafts after launch and tune them as your listings and the rules change, so a person still sends every message.
After the open, you leave a short voice note on who came through. A follow-up for each buyer is drafted in your words and waits in your drafts. You read it, change what you like and send it, so it never takes the final step.
Your note Couple at the open, need a second bathroom, finance sorted, liked the yard.
Draft, waiting for you Thanks for coming through on Saturday. You mentioned a second bathroom: I’ll let you know first when one comes up nearby. Logged in the CRM.
Nothing goes to a buyer until you’ve read it and pressed send.
It learns from follow-ups you’ve already sent, not from a template.
Anything about price comes back to you. The draft leaves it out.
Follow-ups go to buyers who agreed to hear from you, with an easy way to opt out.
The note and the follow-up land in your CRM, against the buyer and the listing.
You check every draft to start with. Checks ease only as it proves itself on your listings.
We scope, build and look after it as part of our AI automation service.
In the writing around each listing that repeats every week. The relationship work, the negotiation and the price stay with your people.
Before anything new goes live, it’s set up, measured and tested on your own jobs.
“The best thing about Paul is not only does he have a deep understanding of AI but also business. As a result he found real opportunities for us, was honest about where AI doesn’t belong, and was clear that the client relationships we’ve built stay ours.”
“He showed us practical, real-world workflows tailored to the way we actually work in real estate, and where AI can genuinely save time and improve efficiency. There was no hype, just a clear understanding of how AI fits alongside the systems and tools we already use.”
A few rules already reach the work AI touches in an agency, so we build them in from the first workflow.
In NSW, the agency agreement includes the agent’s own estimate of the selling price. It must be reasonable, the seller gets evidence that it is, and ads can’t show a lower price or say “offers over”.
Source: Property and Stock Agents Act 2002 (NSW), sections 72A and 73
Marketing emails and texts need the person’s consent first. They must say who they’re from, and an unsubscribe has to be honoured within 5 working days.
Source: Spam Act 2003, as explained by ACMA
Since 1 July 2026, brokering the sale, purchase or transfer of property is regulated by AUSTRAC, and on a brokered sale both the buyer and the seller are your customers.
Source: AUSTRAC, real estate designated services
Australia’s privacy regulator recommends, as best practice, that personal information stays out of publicly available AI tools.
Source: OAIC guidance on commercially available AI products
Already using AI on your jobs? Check one use in about five minutes
Try this yourself this week
Our guide on where AI saves time in real estate, and where it doesn’t covers listing photos, rental ads, AML/CTF and tenant data. For which AI plans keep client details private, read is it safe to put client information into AI tools.
It might be. If your CRM’s AI does the job, lean into it: we set up what you already pay for first. A build only pays where the work runs across the inbox, the portals and the CRM.
No. It drafts, and the draft waits for one of your team to check it and send it. It’s set up so it never takes the final step.
No. It can pull together the comparable sales and show its working, but the estimate is the agent’s, and so is every price a buyer hears.
That’s the fear that holds most real estate agents back: one message that reads like AI can damage a relationship. So before a draft is ready, it checks itself against the listing, and then a person reads it.
Only what the job needs. You wouldn’t give a new junior access to every commission, and it’s the same here. The accounts stay in your agency’s name.
The one you already use. Before anyone quotes a build, we check what your CRM lets us read and write, so you know what’s possible before you commit.
Not necessarily. Every draft is checked to begin with. Where one kind of message has proved itself on your listings, the checks can ease. Price, offers and anything unusual always come to a person.
With the one job that costs the most time, usually the follow-ups or the vendor reports. We prove that on your own listings, then decide on the next.
Book a discovery call. Half an hour, no obligation. I come back with the first step and what it costs.