AMES, IA — Buying a home is a time-consuming process, in part because it requires balancing financial realities with a long checklist of expectations and desires. People care about a solid foundation and certain number of bedrooms. But a property’s curb appeal, neighbors and proximity to work or good schools also matter.
For most house-hunters in the U.S., setting up filters and scrolling listings on Zillow has become a crucial first step.
“Digital real estate platforms like Zillow help people see what’s available, which saves them time by preventing wasted trips to properties that don’t fit their criteria. Even when working with a real estate agent, these platforms often play an important role in the home-buying process,” says Cheng Nie, assistant professor of information systems at Iowa State University.
In a recently published study, Nie and his co-authors highlight how specific features on Zillow influence people’s decisions when making offers and buying houses. Their analysis indicates listings with more “experience attributes” increase the sale price of properties. Photos and descriptions like “upscale bathroom fixtures,” “a sunlit kitchen,” or “an exceptional lake view” fall into this category. They signal the aesthetic and less tangible benefits of a property.
The researchers found experience attributes play an even bigger role in the sale price of homes valued significantly higher or lower than the neighborhood average. Nie says the number of “saves” from clicking the heart symbol on a Zillow post can increase offers and sale prices, as well.
“Before the existence of Zillow and online platforms, people would go to a house, and if they saw a long queue, they would perceive the property as very popular and might offer more. In the digital world, the shortlist info – the number of people who have hit the heart icon – serves this purpose,” explains Nie.
Snapshots in the housing market
To collect their data, the researchers split the U.S. into four regions (Northeast, Midwest, South and West) and identified the smallest, median and largest metros from each. They then randomly selected five houses listed for sale on Zillow in June 2016 from each zip code. Some metros had multiple zip codes.
The researchers collected general property information, like plot size and distance to schools, and data provided by Zillow, including the estimated sale price, number of photos and descriptions.
They looked at the listings again in September 2016 to see which had sold and collected the official sale prices. The researchers also analyzed each property’s tax history and ratings for the seller real estate agents and nearby schools.
A second dataset, collected from the Chicago metro during the winter of 2019-2020, confirmed the researchers’ initial results.
Advice for buyers, sellers, agents
The researchers say real estate platforms can be reliable sources of information for home buyers and offer a reasonable prediction of the property’s value. But they’re not perfect.
“Zillow estimates a sale price for nearly every listing. However, that number is based on the sales of nearby properties. If there are not enough nearby sales, then Zillow does not have enough information to make an accurate prediction about the house,” says Nie.
House hunters may anchor their expectation around the listed price on Zillow, even though the actual value is higher or lower, he explains. Nie recommends checking multiple websites and asking a real estate agent how much a particular house is worth.
For home sellers, the researchers recommend using more experiential features when listing properties. Nie and his co-authors add that the platforms have changed the role of real estate agents but that they can leverage sites like Zillow to deliver better services.
“Instead of merely listing a home as ‘a four-bedroom, three-bathroom property,’ a seller could use experiential features by writing something like, ‘This four-bedroom haven is perfect for families, with a sun-soaked living room ideal for Saturday morning cartoons and a master bathroom that feels like your personal spa,’” says Nie, adding that the latter description taps into the potential buyer’s emotions and experiences.
Hua Sun, associate professor of finance at Iowa State, along with Zhengrui Jiang (Nanjing University); Arun Rai (Georgia State University); and Yuheng Hu (University of Illinois at Chicago) contributed to this study.
Nie’s current research project focuses on iBuyer companies, which use algorithms to quickly buy and sell properties. They often gain a competitive edge over traditional buyers by offering cash deals. During the COVID-19 pandemic, many of these companies shrank or folded, including Zillow’s subsidiary, Zillow Offers. Nie expects a paper out next year.
