REA Group's ChatGPT Property App: What It Means for Buyers

Published on offmarket.now | July 2024

In February 2026, REA Group launched Australia's first ChatGPT property search app. For the first time, Australians can search for homes conversationally inside ChatGPT — no filters, no dropdowns, just natural language.

It is the most visible sign yet that the way Australians find property is about to change fundamentally.

What the app actually does

The integration lets users ask ChatGPT questions like "show me three-bedroom houses in Newtown under $1.5 million" and get back real listings from realestate.com.au. It understands context, follows up on previous queries, and can refine results based on additional criteria like proximity to schools, transport, or parks.

This builds on REA's natural language search rollout in late 2025, which brought conversational search directly to the realestate.com.au website after a 12-month pilot. The ChatGPT app extends this beyond REA's own platform and into a tool that 100 million+ people already use daily.

REA has flagged plans to expand the app to cover rentals, commercial property, and sold property data.

Why this matters for off-market

The shift to conversational search is significant for off-market property for one reason: it changes how buyers express what they want.

Traditional property search forces buyers into a rigid grid: suburb, price range, bedrooms, bathrooms, property type. But buyer intent is rarely that clean. Someone might want "a quiet street near the beach with space for a home office" or "an investment property in a gentrifying suburb with good rental yield."

Off-market properties — which often come with incomplete information, unusual features, or non-standard layouts — are poorly served by filter-based search. Conversational AI handles ambiguity better. It can match a property with three bedrooms and a study to a query for "four bedrooms or three plus an office."

This is the same insight driving postcode-based alert systems. When a buyer subscribes to alerts for a postcode, they're expressing geographic intent. When they type a conversational query, they're expressing lifestyle intent. The platforms that can serve both will win.

The competitive landscape

REA isn't alone. In October 2025, both Homes.com (powered by Microsoft Azure OpenAI) and Realtor.com launched their own AI-driven natural language search tools in the US. CoStar Group, which completed its acquisition of Domain in August 2025, has described its Homes AI as "the most sophisticated vertical AI application in real estate." That technology is expected to flow into Domain's Australian platform.

For smaller platforms, the barrier to entry is lower than it appears. The underlying language models (GPT-4, Claude, Gemini) are available via API. The hard part isn't the AI — it's the data. Having a comprehensive, up-to-date listing database that the AI can search against is the real moat.

This is where off-market aggregators have an advantage. If you're maintaining a comprehensive, geocoded, structured database of off-market listings, you already have the foundation for conversational search. The AI layer is an interface improvement, not a data problem.

What buyers should do

• Try the REA ChatGPT app — it's a genuinely useful way to explore on-market listings conversationally
• Recognise its limits — it only searches REA's own inventory. Off-market properties, by definition, aren't there
• Set up postcode alerts on platforms that aggregate off-market listings. Conversational search is great for exploration; alerts are better for speed
• Combine both approaches — use conversational search to define what you want, then set alerts to catch it the moment it appears

The era of scrolling through paginated search results is ending. The question isn't whether AI will change property search — it's how quickly buyers adapt to the new tools.

---

Browse off-market listings or explore by suburb on offmarket.now.