LOS ALTOS, CA—Prospecting for sellers of real estate is a numbers game and a lot like throwing darts. If a broker talks to enough owners, eventually there will be someone who intends to sell but this requires talking to many who will not. The challenge is to market to the right building owners who actually have a need to sell or buy, filtering out the properties with virtually no chance of selling.
Future Listings are the Bullseye
To address this challenge, a predictive model scores properties and tags them as likely sellers, making it easier for professionals to find properties that are ripe for sale listings. This is the operating platform of ProspectNow, which is an online commercial real estate database.
The database contains information on more than 8 million buildings, 6 million building owners and 30 million commercial tenants nationwide. Through the properties, companies and foreclosure search engines, ProspectNow subscribers can easily locate the phone numbers and mailing addresses of building owners or key LLC contacts. These results also yield comprehensive property information that includes a building’s vital statistics, tenant roster, transaction history and notices of default or foreclosure.
Steve Wayne, founder and CEO of ProspectNow, tells GlobeSt.com:
“When I was in commercial real estate, the most important thing for my business was to get listings. ProspectNow’s predictive analytic platform is a game changer for our customers and will help them focus their marketing efforts on the buildings that are most likely to sell, instead of just randomly marketing to all properties in a given market.”
Back testing results indicate that focusing on the properties ProspectNow has tagged as “likely sellers” or potential “future listings”, may result in as much as twice the probability of a property being listed for sale. ProspectNow’s goal is to eliminate time-consuming database building and duplicated marketing efforts so its customers can spend more time servicing clients and conducting marketing activities that bring results.
“Imagine half the cold calls or direct mail with the same or better results. This is the goal,” Wayne continues to tell GlobeSt.com. “The predictive model is a machine learning algorithm, so with each update, it gets more and more accurate. We have processed millions of properties to date and are excited about what this new capability will mean for our customers.”
The testing that Wayne’s team performed indicated that there is a correlation between a property that sells and particular events and conditions. About a year ago, the team set out to see if it could build a predictive algorithm that would take into account the way a property looked before it sold, and then apply these similar characteristics to properties that have yet to sell to filter out the non-sellers.
For example, if a property was just purchased 10 days ago, it’s unlikely that the owner is going to sell it any time soon. Veteran real estate brokers know there are lots of different triggers as to why an individual might sell. But, Wayne’s team wanted to stop guessing and prove it using hard data and analytics.
The ProspectNow team started running its model each week for a year. After the model was trained, the team then looked at a group of properties tagged as “likely sellers”. They continued running the model and updating scores on properties each week. The goal was to determine if the list of properties the model generated the previous 10 months actually resulted in a higher percentage of properties sold. In some cases, more than twice the number of properties sold in the category tagged as “likely sellers”.
ProspectNow clients simply login, search and select the check box “likely sellers” in the search to get the results for the properties in a particular market.
Original Source: Lisa Brown/GlobeSt.com Predictive Model Identifies Future Listings