Searching For Your Dream Home Just Got a Whole Lot Smarter

Searching For Your Dream Home Just Got a Whole Lot Smarter

For decades, finding your perfect home online has meant scrolling through listings and hoping the agent remembered to mention the wood floors, the sun-soaked kitchen, or the home gym tucked behind the garage. That era is over.

 

realestate.co.nz has unveiled a major shift in how Kiwis search for homes, an AI-powered image search tool that is a first for any property website in New Zealand. Instead of relying on written descriptions, buyers and renters can now search for the features that genuinely matter to them, a weatherboard exterior, wooden flooring, home gym, or a swimming pool, with results surfaced directly from the property's photos. "Buying a home is an emotional decision. People respond to how a space looks and feels - Sarah Wood, CE, realestate.co.nz

 

Two layers of intelligence

Behind the scenes, the technology brings together two distinct AI capabilities. First, every image uploaded to the platform, stored within Snowflake's AI data cloud, is analysed to identify key features: room type, interior style, flooring, standout design elements like custom shelving, and even more subjective qualities such as a great view or a shady garden.

 

The second layer is about how that data is used. The platform's search functionality has been upgraded to interpret both exact matches and related concepts, helping users discover properties they may not have initially thought to look for, broadening the journey in helpful and often surprising ways.

 

Thirty years in, and still redefining the search

The launch is significant not just technically, but symbolically. Established before Google, realestate.co.nz has been helping Kiwis buy, sell, and rent property since 1996, making it New Zealand's longest-standing property website. This milestone arrives in its 30th year.

 

CEO Sarah Wood says the innovation is a direct expression of the organisation's core mission.

In our 30-year history, we've developed a deep understanding of how people connect with property," she explains. "Property searches have traditionally depended on how accurately a home is described. But this tool connects people with homes based on what they're genuinely drawn to, whether that's a swimming pool, home gym, wine cellar, or walk-in wardrobe.

Wood also highlights a practical upshot for buyers and renters: less time fiddling with filters, more time discovering homes that actually match what they want. For sellers and agents, it raises the stakes on photography, high-quality images that showcase a property's best or most unique features now directly power search results.

Thirty years ago, we brought property searching online. Now we're redefining it again" - Sarah Wood, CEO, realestate.co.nz

 

The data foundation that makes it possible

Tony Shaw, Country Manager New Zealand for Snowflake, says the tool demonstrates what AI can achieve when the underlying data infrastructure is right.

realestate.co.nz manages vast amounts of data including photos, maps, addresses, and listings. By leveraging Snowflake's Cortex AI, they're using this wealth of data to personalise the customer experience with high accuracy and at significant scale and this benefits all New Zealanders buying, selling, or leasing property.

The AI image search joins a growing list of realestate.co.nz innovations, including Tāhū, the AI assistant that simplified the selling process,  and AI-powered floor plan surfacing.

 

The tool is now live in beta on the realestate.co.nz app. Kiwi property seekers are invited to be part of searching history and share their feedback.  

 

To find out more visit realestate.co.nz

 

Comments

Related Blog Posts

Our Members

Our members include advertisers, agencies, ad tech companies and media owners, so you're in the right place to discover the very best of digital advertising.
Index_Exchange_Logo_Full_Stone
MetserviceHorizontalWhite_a0aa5fd9-5ae6-4a79-a1b3-7097e5c5858b
sky-logo-white-rgb
WHITE Logo Snowflake-VML
Are Media Logo White2
MIntHC_Logo_WHITE (1)