Justin Mares today announced his new roommate matching software “RoommateFit” which he’s pitching to college campuses to aid in their dorm room assignments. The University of Pittsburgh Business student’s system will ask 42 questions, covering 6 distinct traits, in order to better match like-minded students with one another.
I recall talking with a co-worker in 2001 while working at a large, all-inclusive, student community here in Gainesville who realized the need for exactly such software. “So wait, you mean there’s no such thing as 3rd party roommate matching software out there for college communities?”, he asked me incredulously. “Not really” I said, “most people make their own proprietary system but the whole thing takes a back seat to other business aspects of the operation.” He resolved right then and there to make the first, one-size-fits-all roommate matching software for college campuses and stand-alone communities.
It’s been a decade and I haven’t heard tell of him, or his software, yet.
And that’s probably because it’s not as easy a thing to do as one may initially think. What really matters when matching roommates? It’s not like they’re dating… they just need to do a bit better than not kill each other. So, preferred bedtimes are the big deal? Drinking? Hobbies? Majors? Affiliations like Sororities or clubs? Music choices? How’s the hipster vegan from NYC going to get along with the Agriculture major from Texas who believes Bacon isn’t just a food group but a full-blown religion?
How do you correctly weight the answers to these questions and which factors hide the buried fault lines of potential friendship or ever-lasting war? That’s the problem most campuses and Property Management groups have when it comes to finding “Martin” his new “Lewis.”
If you’re a new college student and are getting an assigned roommate I have a sobering truth for you: That ‘random roommate’ is way more RANDOM than you may actually think.
Many times, and due to the actual calamity of placing/assigning units and coordinating dates/furniture/vendors, etc… roommate matching can, and is often, done based entirely on gender alone: “I got a single guy here for a 4/4, you have a space open with three other guys?” says the Leasing Manager to the Property Manager – “Yeah, done and done. Placed in 1010B, next?”More on this exact process later…
This is where Mares’ software really becomes appealing; if you can sort people to be placed together based on similar interests, reduce immediate roommate conflicts, and speed up the process so the office staff is free to focus on preparing your unit correctly for move-in, everyone wins.
The big question remaining to be seen is this: Will Mares’ system create better matches than random assignment? Significantly better as to justify the cost of his annual service? And how do you even measure such a thing? What’s a successful placement? Is this system better than letting students find their own roommates on Facebook?
Those questions are why most groups don’t see money in attempting to ensure roommates hit it off… “random” or limited review of roommate preferences has worked thus far, therefore, why spend the money?
Robert H Myers 
Leave a Reply