According to Wikipedia, “MLS has never taken hold in Manhattan.” But the technical future of the Multiple Listing Service may be just what the over-13,000-member Real Estate Board of New York (REBNY) has been exploring for years with its listing system, the RLS.
From time to time in the industry, someone will opine that the industry should do away with a centralized MLS and move to a “peer-to-peer” system. Until recent years, the way the system worked in Manhattan was just that— a “peer-to-peer” system. Brokers would enter their listings into their own database – some hosted by a technology vendor, some self-hosted – and on a daily basis, those databases would synchronize with each other using the market-wide “R.O.L.E.X.” standard (sort of like the RETS “data dictionary”). These data payloads were sent in delimited files, using the old FTP protocol. This system had issues with the divergent data standards of each participating listing system, synchronization challenges, and duplicate listings being created. It was very difficult to address some of these issues because, with the peer to peer system, there was no one authoritative source for listing content. REBNY staff was often on the phone with multiple vendors trying to figure out what a listing looked like in the database, where it had been entered, and why it was missing or incorrect in other systems. “Peer-to-peer” was simply not a panacea for broker listing exchange.
In 2010 Clareity was called in to help create an improved system for the RLS. The new system would allow brokers to continue to maintain the model where they could use any of seven “front-end” systems, all of which would communicate with a common “back-end” repository using industry-standard RETS. In short, the new system maintained choice for each firm – they were not relegated to one “front-end” system, but could choose to keep their incumbent system or move to a new one. REBNY contracted for the “back end” to be provided by Stratus Data Systems in late summer of 2011. REBNY hired nationally recognized MLS Executive Freddy Sarabia as the RLS’s first Executive Director, and created a separate RLS Board of Directors, chaired by Warburg Realty CIO Steve Goldschmidt, to oversee its operations. After the back-end was built and front-end vendors coordinated – and a parallel period where both systems ran at the same time –the final transition to the new system took place in September of 2013. Starting in 2015, the wider real estate industry started talking about splitting MLS into “front-end” and “back-end” via RPR’s “Advanced Multi-List Project” (AMP) – but REBNY had been proving out the model for many years, and this model has been working in its most current form for more than two years. Like they say, “Everything old is new again!”
I asked friends at the RLS, including REBNY General Counsel Claude Szyfer, a partner at Stroock & Stroock & Lavan, what lessons the industry and RPR could learn from the RLS’s experience. What I heard back can be summed up as follows:
- The separation of listing system front- and back ends has allowed the creation of a system that fosters competition and innovation in the market for technology services. Ensuring brokers have a choice of front-ends makes brokers happy.
- If MLS systems are designed to scale down to this level, MLS vendors will be able to compete for business on the broker level and not just on the MLS level. We may have to come up with a new name for them, as the “MLS system” category may fold into the “broker system” category.
- The separation of front-end from back-end will allows RPR to just focus on getting the back-end perfect— they don’t have to worry about coming out with a new front-end system every year. While it took a little longer than anticipated for RLS to launch fully and run solely on the Stratus system, there have been no technical issues of note. RPR should be patient and focus on getting the distribution engine perfect. Getting it right the first time is better than rushing out with a product that does not have great functionality. “You never get a second chance to make a first impression.”
- Communicate with your constituents. Get them involved at the inception, work with them, and find out what they want. Talking to brokers is an incredibly enlightening process.
It does seem as though RPR’s AMP is going to take a lot more work to create than it took to implement REBNY’s RLS. After all, making that same model work at the national level involves a huge number of local fields and business rules, as well as the divergent interests of many associations, MLSs, brokers, agents, and a huge number of software vendors that will want to work with that back-end. Also, one of the benefits desired for AMP would be the ability for MLSs, or even brokers, to easily “swap” one system for another. That takes a transmission standard for far more than just the listing data – as anyone who has gone through a system transition or put a parallel system in is well aware – and we don’t even have RETS data standards for much of that data yet.
All of that said, seeing how the REBNY RLS works can provide a valuable basis for understanding how, at the local market level, the MLS “front-end” / “back-end” split that AMP is based on can work and, though it doesn’t have to provide front-end of choice for brokers like they do in New York City, it certainly can. The industry is watching as RPR attempts to scale and expand upon that model with AMP. Meanwhile, some markets – especially those that don’t have additional problems to solve like “overlapping market disorder,” may look at the RLS model and ask, “Why not do something like REBNY’s RLS in our own market?”
Share this post: