Lens is the latest tool from the tech-forward brokerage Compass, announced Wednesday.
The enhanced search functionality empowers agents to demonstrate to clients what a room might look like after a major renovation and, thus, how value is affected.
In what is likely one of the closest application of actual machine learning in the industry, Lens scours local multiple listing services, as well as the Compass listing network, to pull homes with similar “before and after” renovations.
Lens is rooted in the object and pattern identification abilities of what’s called computer vision, a form of machine learning that assembles and processes characteristics of images. (Computer vision can be used in highly advanced scientific scenarios.)
The technology is commonly seen in Google Photos’ ability to organize images of your dog or categorize pictures with surfers.
Kitchens are the most obvious example because they’re often the most important to buyers, they have a measurable impact on price, and as of now, they’re the “easiest” for the software to identify.
Major appliances, such as stoves and refrigerators, tend to standout. Amenities like a breakfast bar or a backsplash’s subway tile finish can also be easily pinpointed in an image search.
Using a slider interface (picture a horizontal window shade), Lens’ search results allow the user to reveal the look of the updated room against its previous iteration. The intent is to demonstrate the impact the renovation has on sale price.
Compass stated in the press release that before Lens, “an experienced agent would walk through a home and determine which improvements should be made.”
To some extent, Lens can be looked at as an upsell for Compass Concierge, the brand’s proprietary program that lends money to sellers and manages all aspects of upgrades and market-ready maintenance.
The idea stems from the company’s selling mantra of “Price, Prep, and Promotion.”
Compass, which refers to agents as its customers, had members of the team behind Lens attend listing presentations. They extensively surveyed how buyers and sellers pitched business, according to Greg Mushen, a senior product manager and one of 12 people on Compass’s AI team.
“We recorded visits and spent hours deconstructing them,” he said.
Where they saw the most need for improvement was in the “prep” stage. How a home presents to the market is crucial to its ability to compete, especially in suburban markets popular with tract housing developers in which floorplans and finishes are often replicated. A unique interior design helps drive interest.
Lens isn’t consumer facing — it’s an agent tool for them to show leads. Using a room filter, the software recalls every market listing with an appropriate renovation story and the imagery to back it up.
Living spaces are getting easier for Lens to pick out. Couches, large windows, fireplaces and open space are helpful identifiers.
Compass agents can also use pictures to search for a listing.
Buyer’s agents, for example, can use the tool while on tour (where appropriate these days) by uploading a photo taken onsite to search for nearby comparable listings that may have had a recent renovation. This helps demonstrate an agent’s tech savviness to the buyer, and helps the buyer better understand their offer position by seeing how a comp with a more modern interior changes it.
Mushen said that exteriors are likely to be the next major home component to leverage the image embedding technology.
Have a technology product you would like to discuss? Email Craig Rowe
Craig C. Rowe started in commercial real estate at the dawn of the dot-com boom, helping an array of commercial real estate companies fortify their online presence and analyze internal software decisions. He now helps agents with technology decisions and marketing through reviewing software and tech for Inman.