Project Overview
Australia’s real estate portals are the most profitable in the world, with over $900m in annual revenue servicing 15m consumers, however, their advertising model remains tied to the old newspaper advertising model - holding back real consumer-focused innovation.
Homes.com.au are seeking to disrupt the Australian property advertising market by developing new residential and commercial property portals that utilise a consumer-first data and artificial intelligence driven listing platform. The aim is to deliver a highly personalised real estate discovery and searching experience that provides a unique and enjoyable way to explore property with true personalisation to help the consumer to stop searching and start finding.
Project Commissioner
Traffic X on behalf of Homes.com.au
Project Creator
Team
Andrew Begg - CEO / Creative Director
Matt Smith - Project Lead / Digital Director
Paul Simos - Executive General Manager
Dan Harper - Chief Technology Officer
Justin Lilburne - Product Manager
Jared Bon - Design Lead
Tony dos Santos - Senior UX Designer
Rigel Maple - UX Designer
Benny Thomas - Engineering Team Lead
Andrew Begg - CEO / Creative Director
Matt Smith - Project Lead / Digital Director
Paul Simos - Executive General Manager
Dan Harper - Chief Technology Officer
Justin Lilburne - Product Manager
Jared Bon - Design Lead
Tony dos Santos - Senior UX Designer
Rigel Maple - UX Designer
Benny Thomas - Engineering Team Lead
Daniel Gil - Data Scientist
Damon Rasheed - AI Data Scientist
Theo Voronov - AI Data Scientist
Todd Galloway - Front-end Engineer
Chris Chan - Front-end Technical
Nicholas Terrell - Senior Engineer
Jordan Schuringa - Senior Engineer
Sachi Karunaratne - Senior Engineer
Khoa Le - Software Engineer
Danyal Lakzian - Software Engineer
Project Brief
The brief from Homes.com.au called for Traffic X to design a real estate portal that is different from all others in the market, whilst not looking different - it needed to feel familiar to the consumer.
The portal needed to act like an invisible buyer’s advocate. Utilising engineered intelligence to assist the home buyer, identifying their needs and wants based on viewing, search habits and preferences - much like Spotify and Netflix do for entertainment purposes, and present them with options they would not otherwise have found.
The aim is to leverage the data intelligence to allow the average property buyer to find their ideal property within weeks, not months - providing a unique and enjoyable way to explore property with true personalisation to help them stop searching and start finding.
Project Need
We all love how Netflix, Spotify and YouTube’s algorithm suggests more content to us, but why not real estate? The great Australian past time of searching, browsing and buying real estate is stuck in a ‘he who pays the most gets the visibility’ model. This isn’t user-focused, it is vendor/agent focused, driving up the cost of advertising unfairly, without any benefit for the end consumer using the platform or the vendor and agent.
By switching the focus to the consumer it opens up many possibilities for changing the way we search for property. It allows us to mix up results, allowing users to see more of what they want and less of what they don’t. This model is achieved through many different innovations, but it starts with data.
Traffic X built an Artificial Intelligent driven search engine for Homes.com.au that learns more about each listing, suburb and user onsite behaviour than ever before. Our deep learning neural network is able to process data from open source, corporate and government data sets, and combine and learn with our real estate listings that we have processed and enriched with natural language processing, image recognition, comparative modelling and property life stage segmentation.
The end result is set of search results that is unique to each user, a set of results that learns the more you use it, delivering better results each time your search and discover more real estate.
User Experience
Extensive user testing was conducted exploring all aspects of the platform from search flow, property listing cards, listing optimisation and conversion to brand messaging and placement - maximising the user first, not the traditional advertiser first approach.
Particular focus was placed on the UX for the non-standard real estate portal features, those that would separate Homes.com.au from it competitors. Investigations focused on features that could be developed based on the A.I data enrichment capabilities - what would be useful to a customer and what wouldn’t. The A.I driven search engine’s successful learning program enabled Traffic X to deliver product features that we're able to take advantage of the data enrichment. These included;
- Placelists - like a playlist for music, but for real estate and property;
- Similar suburbs - using market intelligence we find similar not just surrounding suburbs;
- Feature tags - preference your wants and needs and see more of what you want;
- More search possibilities - more options, easier to search and better results;
- Targeted relevance - personalised search delivering hot leads not time wasters;
- Property Hub - one central location to manage the entire property experience.
Project Marketing
The Homes.com.au brand has only just begun its marketing phase, attracting new customers and vendors and agents alike. To truly succeed it must attract both, in large numbers.
To do that, the opportunities afforded the brand through going to market with a truly innovative product and offering customers and vendors a differentiated product from the competition set a foundation for the brand to thrive in the marketplace. Combined with a standout, fun and personalised brand and campaign sets forth the portal for success in the future.
Project Privacy
When building a platform based on A.I and the use of data, privacy was the first and second considerations - it has to be. Data is primarily used in two ways on the platform; quarantined personal data and anonymised mass data.
Quarantined personal data is protected and only accessible and utilised by direct input from an individual. The data and inputs of a user are available only by that individual user, never shared.
Anonymised mass data is when we utilise behaviours and patterns by the customers, but all self-identifying characteristics are removed. This allows for behaviours and inputs to be compared and contrasted with other users and other data sources, with the A.I identify patterns and opportunities.
Digital - New Service or Application
This can be any new service or application from a start-up to an industry leader. It’s not just bells and whistles we're after but true innovation, exceeding expectations and filling a void that had previously been open.
More Details