While most home sellers and buyers still choose their real estate agent by word-of-mouth, there are an increasing number of tools available to find agents online. These tools can be structured around consumer ratings and reviews of agents, or actual agent performance as measured by a range of statistics.
NeighborCity.com, an online brokerage and property-search site owned and operated by licensed brokerage company American Home Realty Network, has built an algorithm for agent performance that factors in the agent’s location, the number of bedrooms, the price range the agent specializes in, how the sale prices of the homes the agent sold differs from the listing prices, how long those properties remained on the market relative to comparable properties, and the agent’s client volume.
While the site includes comments about agents from consumers, those reviews are not included in its algorithm.
Within the next few months, NeighborCity has plans to display more detailed statistical information on individual agents that will be accessible to its registered users.
"Opinions come from a certain context that is lost," said Jonathan Cardella, NeighborCity’s CEO. "Numbers don’t lie." Before founding NeighborCity in June 2007, Cardella founded Ski West Inc., an online vacation-rental booking company that later sold to Overstock.com.
Cardella posits that perhaps 10 percent of real estate agents do 90 percent of the business in the industry. NeighborCity seeks out co-brokerage agreements with the top 10 percent.
"Our theory is that for each property there’s maybe five to eight agents who are really qualified (to handle it)" and the system finds those agents for consumers, the CEO said.
The company has signed about 20,000 agreements so far out of 365,000 agents listed on its site, most of them focused in 20 metro areas heavily represented on the East Coast, West Coast, Texas and some Midwestern states, Cardella said. The company will expand to Illinois next.
About half of NeighborCity.com’s users come to the site for agent searches, and that number is growing quickly, Cardella said, adding that the other half of the site’s visitors are there for property searches.
Consumers can enter a location and the price range of the home they are looking for to view a list of ranked agents to choose from.
The company contacts the agent and, if one is not in place already, offers the agent a co-brokerage agreement in which that agent would typically get 70 percent of the commission from the final transaction, Cardella said. NeighborCity would get the rest in return for the referred business. Agents do not pay for their rankings and are listed regardless of whether they choose to work with the company.
As it stands now, the site gives very little explanation for why agents merit their rankings. But the company plans to roll out new features to appease consumers’ hunger for transparent and reliable data in the information-drenched chaos of the Internet.
Location is weighted most heavily in the algorithm, followed by price range — and the new features will shed more light on agent activity and performance. Among the planned changes: agent profiles that have been claimed can be populated with an agent’s Twitter and blog feeds, for example, and the profiles can also display updates when the agent gets a listing or completes a sale. …CONTINUED
All profiles will display agent statistics — both on the buy side and sell side — and will help visitors to make sense of the numbers by showing how a selected agent stacks up against an average agent.
On the buy side, stats include median value of homes purchased, the local market average home value, the kind of home the agent specializes in, the number of buyers who purchased a home in the past year vs. the market average, the reduction in price from a list price, and the market average of reductions for comparable properties.
On the sell side, stats include the number of sellers the agent worked with in the past year who sold homes, the market average for number of sellers, the average listing price of homes the agent sold, the average listing price of comparable properties, the average number of days on the market for the agent’s properties, and the average days on the market for comparable properties.
"Are they productive? Do they provide value for their clients? Do they always undersell? There’s a difference between an agent who performs well for their clients and an agent who just gets through it," Cardella said. Because they aren’t based on just one transaction, he said, these numbers will bear out whether agents are performing in the best interests of their clients overall.
Data for the algorithm comes from local multiple listing services. Performance metrics, specifically, are derived from Virtual Office Web site (VOW) feeds that users can access only by logging in to the site. Rankings don’t remain static, but change as transactions occur and these sources update. The database also incorporates data from transactions the company has co-brokered.
MLSs typically provide Internet Data Exchange information, which is listing information shared between brokers. VOW data, while richer than IDX data, can be provided only to registered site users and can require additional technology expenditures to implement.
Some brokers question whether the investment is worth it, Cardella said, especially for tech-challenged brokers who may be struggling with basic property search on their sites or are not sure the VOW data will actually help them get more clients.
And real estate professionals’ opinions of agent ratings and rankings sites differ from the frightened to the cautiously optimistic. Most, however, acknowledge that agent-ratings systems are not going away.
"It isn’t right or wrong; it’s what the consumer is demanding," said Nobu Hata, a Realtor at Edina Realty, commenting on an article by Inman News columnist Kris Berg that questions the validity of agent ratings.
Currently, more homebuyers and sellers find their Realtor through a referral from someone they know than by other means. According to the 2009 National Association of Realtors Profile of Home Buyers and Sellers, about 44 percent of buyers and 64 percent of sellers participating in an NAR survey found their agent through a friend or relative. Ten percent of buyers found their agent online, compared with 3 percent of sellers, according to the survey report.
In the 2008 Houston Association of Realtors (HAR) Survey of Home Sellers and Buyers, more than 50 percent of homebuyers and sellers found their Realtor through a friend, neighbor or co-worker. Nevertheless, a substantial majority, 81 percent of sellers and 75 percent of buyers, said they would find it beneficial or very beneficial to access ratings from a Realtor’s previous clients.
In recognition of these preferences, the National Association of Realtors recently announced funding for a rating site called "Rate Your Experience" from the Peoria Area Association of Realtors. Rate Your Experience will use consumer reviews to evaluate all service providers involved in a real estate transaction, including brokerages, real estate licensees, lenders and inspectors.
HAR also recently launched its own ratings system, the Client Experience Rating program. …CONTINUED
Several agent-ratings sites filter by location first. "There are so many advantages to a local agent and yet so many people take a friend referral that doesn’t take into account empirical data," Cardella said.
Some critics say agent-ratings systems are attempting to quantify an essentially subjective experience.
"Numbers of homes sold, average sale price, and even years in business — those things are science — and science has no home in our world, where pleasing the customer is an art form," Berg said in her article. The only rating system that matters, she said, is a personal referral from a customer.
Nevertheless, despite critics’ fears that such sites will be, at best, misleading, supporters say the sites could actually help qualified agents to shine.
"We want to bring out the diamonds in the rough who might not have as much volume but are really focused in one area," Cardella said.
Agent-ratings sites often allow agents to respond to negative feedback or correct misconceptions or errors, too, or even to opt out from participation.
Cardella said that NeighborCity works to "incubate the buyer" by exposing them to properties in searches at the site. "When we send clients to agents, they’re prequalified. They already know what home they want. The agents don’t have to drive them around," he said.
The AgentMatch system also contains an "intelligent" routing system for referrals that tracks how busy agents are and moves on to the next best agent on the list if an agent rejects a client. Some criticize NeighborCity’s approach as too standardized.
"Doesn’t this mean that every buyer who runs a search for a $600,000, three-bed/two-bath home in San Jose is going to receive the exact same agent rankings/recommendations? How can an agent be the best/most qualified for all buyers? Don’t buyers have different needs/preferences?" questioned Michael Erdman, president and founder of AgentsCompared.com, in a comment to a blog post about NeighborCity.
Responding to that comment, Cardella acknowledged that two buyers with the same criteria will indeed receive the same results from AgentMatch, but also pointed out that NeighborCity’s system updates daily and "since we use objective, empirical data, we return the most experienced real estate agent for each request."
Later, he said, "Some might call it cookie-cutter, but I call it reproducible, and that’s a good thing."
"One thing we should learn from the economic downturn is that we need to think about the consumers, because if the consumer is gone, everything’s gone. That means transparency and good results," he added.
Jonathan Cardella will co-host a session titled "Breaking Data Taboos for Fun and Profit" at the Real Estate Connect conference in New York City, which runs from Jan. 13-15, 2010.
What’s your opinion? Leave your comments below or send a letter to the editor.