Realtors Property Resource® (RPR®) is 100% owned and operated subsidiary of the National Association of REALTORS® and is offered as a member benefit of NAR. RPR® actively collects listing data from 96% of the MLS’s in the United States and marries that with public record data demographic data and neighborhood data to provide analysis and reporting tools to help REALTORS® service consumers in the buying and selling of their homes. RPR® is an invaluable resource for REALTORS®, and we were excited that they were able to provide our hackathon participants with a special hackathon-only version of their Sandbox environment for developers.
Our friends at RPR® had a bunch of awesome ideas about how their data, combined with artificial intelligence and machine learning, could be used to improve how our members buy and sell homes – the ideas are so good, I’m going to let RPR® give their ideas directly instead of summarizing it from the emails we had back and forth during the hackathon process.
Neighborhood Turnover Rate: For example comparing count of actual SFR properties in a geo location zip census tracts etc and the number of properties closed in a month (run multiple counts month by month for the last year). Also compared the average/median sale prices over time to track trends.
– This could help Realtors and Brokerages identify neighborhoods in which they can focus their marketing efforts for a higher ROI.
– This can also help Brokerages measure possible revenue based on sales volume (sales price volume).
Correlation between distressed properties and the last sale date or sale price: For example search for SFR properties with a distressed date (YTD) in the city of San Francisco. Compare the distressed date to the sale date track a pattern and also analyze the sold price.
– This could help Realtors locate areas and price points that are about to be impacted by foreclosures.
– This could also provide Lending and Government institutions with a timeline of when the purchases occurred so they can be linked to changes in the job market interest rates lending options at the time or lack of market price bubble etc…
A couple of our teams used RPR® data in their final projects, using the data from fifteen million public records across several states in order to analyze trends in public records and property listing data. You’ll notice a theme of several of our sponsors is to collect data from multiple places and create a more streamlined system for displaying it; artificial intelligence and machine learning represent the next step in the process, where the streamlined systems can be analyzed in order to facilitate an even easier buying/selling process for agents and their clients.
Benutech was another one of this year’s API sponsors, allowing our hackathon participants to access their data solutions to use within the participants’ projects. Benutech’s products allow for the easy sharing of multiple data platforms, allowing for seamless integration of the data and a one-stop solution for industry professionals to find out important property information when buying/selling homes.
Benutech tracks a variety of sources – from divorce filings to property tax defaults – in order to help agents generate leads. By tracking all these variables, instead of just one, real estate agents can see all the possible leads in their area instead of scrambling to put together data from different sources. Benutech’s strengths lie in their partnerships with other data providers – including some of our other hackathon sponsors – which not only widens the scope of their searches, but allows for sources to be easier to track and use. One of the biggest complaints we hear in the lab is that all this technology and data comes from so many sources that it’s hard to keep track of, which is why we were interested in working with Benutech. Their collaborations allow them to lead the way for all data providers to work with each other, instead of against. That was also the spirit of our hackathon – using artificial intelligence and machine learning to make real estate more streamlined, and getting groups together to collaborate to create innovative new solutions to the complaints we hear about from Realtors every day when it comes down to adopting technology into their business practices.
We thank Benutech for their sponsorship, and know that they were instrumental in making our first hackathon a success for us and our participants.
TLCengine is a company specializing in the “True Lifestyle Cost” of home buying. Beyond just mortgage, taxes, and insurance, TLCengine looks at over two dozen cost variables associated with owning a home. These factors include average utilities, commuting costs, car insurance rates, and more. Using predictive analytics, TLCengine can give buyers a highly accurate breakdown of how much house they can afford. The traditional way of looking at homeownership costs (the mortgage, taxes, and insurance mentioned earlier) only account for 50% of the costs associated with living in a home – and TLCengine accounts for this in their predictive modeling.
TLCengine was one of this year’s hackathon sponsors, and their data was instrumental in helping our teams who were looking at machine learning/artificial intelligence to help consumers and agents make better home buying decisions. This type of data is uniquely suited for helping clients make smart decisions with their money, beyond just the initial purchase of a home, and we thank them for letting our hackathon participants use their API this year, and wish them luck in launching TRUE, their new product for agents and brokers, this year!
You might already be familiar with Foursquare, which started in 2009 as social-networking-meets-Yelp-reviews. Users interacted with their environment via “check ins,” and competed against their friends to dominate their neighborhoods – and to learn more about local businesses and services. While Foursquare’s app has changed over the years, its database of points-of-interest has grown, with over 105 million documented places in their database. That’s a lot of useful location information, and we were excited that Foursquare offered our Hackathon participants access to their Places API.
Foursquare, and other geo-tagging and venue search services, prove pivotal to the home-buying experience. Buyers want to know about the neighborhood they’re moving to, and the best way to find that out is from the people who frequent the area. Knowing their opinions – and knowing how often they drop by that local coffee house you’re interested in – can help gauge opinions and sway buyers to (or away) from certain neighborhoods. Using artificial intelligence and machine learning, a great real estate application for this could be to get to know a buyer’s current taste (by analyzing their Foursquare profiles) and using that information to find the best neighborhood for them. The stress of moving long distances can easily be reduced if getting to know your new location could be based on years of check-ins to venues in your old neighborhood.
Thanks again to Foursquare for letting our Hackathon participants dive in to your Places data for our event!