Friday, July 24, 2009

MLS Hell: Coping with Data Normalization in an Abnormal Industry

I am so excited to be part of Inman's Real Estate Connect San Francisco 2009 ConnectTech Workshop. Specifically, I will be on the panel titled "MLS Hell: Coping with Data Normalization in an Abnormal Industry". The full panel card is as follows:





David Harris, Director of Data Management, eNeighborhoods


Mark Tepper, Vice President of Business Development & Sales, Homefinder.com

This is just one of a full day of technical focused sessions on August 5th. The entire workshop is moderated by my good friend:


Dan Woolley, cofounder of W&R Studios, creators of Dwellicious.

It's shaping up to be a great conference. Mark, Chris and I have all compared notes to ensure you get the most comprehensive and info-packed presentation on all the issues and solutions the three of us have encountered when interacting with the hundreds of MLSs throughout the country. But don't let the technical nature of the session scare you off, I think the topic can appeal to a wide variety of participants:
IDX/MLS data vendors will obviously be able to walk away with tips and lessons learned that they can implement in their operation.

Brokers and Agents can observe what goes into MLS data implementations and get some great tips on how to evaluate your solution/vendor rates against some of these concepts.

MLS Operators can get insight into the how "the other half" operates and can get great ideas on how to design your data solution to best benefit your membership.

Other industry operators who may not currently work directly with MLS data but interact with listing data from other sources can see first hand if the benefits from direct access outweight the investment necessary.

You'll have to come to the session to get all the good ideas/information we will be serving up, but I thought I would offer a sample here. Recently, the twitterverse had a brief discussion on a geographical representation of the MLS network.





I chimed in with our map:


But the idea stuck with me so I wanted to do a little more work. Microsoft Map Point turned out to be the best tool for mapping a large amount of data. So I first mapped our MLS network (570 and growing) with dot sizes based on the number of active listings (sorry, not interactive... yet):



And then decided to map all of the actual listings (4.3 million and growing) grouped by county. Think of it as a green heat map, the darker the green, the more listings:



If this is cool stuff, come to our session at Inman to get much, much, more...