Alley Guide, a real estate tech startup founded by former Google Maps Taiwan head Cheng Yi-huan, officially launched its "Alley Guide" home-finding map service today (February 4th), focusing on "lifestyle-first" solutions. This product attempts to break away from the traditional logic of rigidly filtering by administrative district and total price, instead centering on "commuting time" and "map experience." It also previews plans to incorporate more generative AI and other technologies in the future, making finding a house as intuitive as using navigation.
It not only calculates distance, but also creates its own maps to calculate "commuting time".
Unlike traditional real estate brokerage platforms that mostly use "straight-line distance" to determine proximity, Xiangdao uses its self-developed spatial computing engine to introduce the concept of "isochrones".
Simply put, users no longer need to guess "how long it takes to get to work from here" based on their feelings. Instead, they can directly input their commuting method (such as MRT or driving) and acceptable time range. The system will then draw an irregular "living circle" on the map, and the properties within that circle are the ones that truly meet your daily commuting needs.
To achieve a smooth experience, the guide abandoned the approach of simply connecting to existing map APIs. Instead, it adopted WebGL technology and a vector map (Vector Tiles) architecture to speed up map loading and avoid frequent loading status updates that would affect the user experience.
It is worth noting that the guide emphasizes that its map data is drawn by itself, rather than directly using Google Maps or third-party map data. This allows it to optimize the underlying data more flexibly, making the map operation not only smooth but also interactive in real time, solving the pain point of having many windows open and fragmented information when searching for a house in the past.
Solving the pain point of duplicate listings, holding 10 "hidden" property listings
In terms of property listings, the guide attempts to address two issues: "information noise reduction" and "exclusivity".
First, to address the issue of the same property being repeatedly listed by different real estate agents, resulting in a dense and cluttered map, the guide uses an algorithm to integrate the listings, allowing users to view all the listing information for the property in a single window.
Secondly, LaneDirector announced that it has partnered with nearly ten of Taiwan's top real estate agencies, boasting over 10 exclusive listings on its platform. Most of these listings are firsthand information from partner agencies that haven't yet uploaded them to major platforms, making this LaneDirector's biggest content advantage in its initial phase.
Future plans: AI voice-based property search, WebVR, and cloud expansion.
Although the service has just been launched, Zheng Yihuan also revealed her future technical plans.
• AI applications:The team is currently focusing on using generative AI to help real estate agents automatically generate listing information, reducing paperwork burden. In the future, they plan to integrate AI into the front end, allowing consumers to find houses directly through methods such as "verbal description" (e.g., "Help me find a house with a large balcony that is 20 minutes away from Xinyi District"), making the search experience more intuitive.
• Immersive experience:Considering the immersive experience of viewing properties, we are also considering introducing WebVR technology in the future, allowing users to conduct more in-depth online property viewings in their browsers, and even immerse themselves in property viewings or experience the surrounding environment through virtual reality headsets.
• New services and overseas markets:In addition to the current sale of used houses, the guide will also include information on new construction projects and developers, and is considering developing rental services. It is even assessing the possibility of entering overseas markets, especially since there is still huge demand for housing transactions in markets such as Japan, Thailand and the United States. Therefore, Cheng Yi-huan believes that there is a certain development opportunity in overseas markets, but at this stage, the focus is mainly on the Taiwan market.
In terms of infrastructure, the current map navigation service is built on the Google Cloud platform to ensure the stability of map operations under high traffic. As future features are added (such as AI computing models, VR video streaming, etc.), a multi-cloud architecture may be adopted.
