MLS reconsidered

Four years ago, I wrote a post outlining some of the problems I saw in the MLS world. It was a complaint born out of experience working with brokers and software developers struggling with a siloed, politically irregular system.

Because I wanted to be constructive, I followed that post with another in which I outlined an idea for addressing some of the problems I complained about. The idea was something I called the “data socket” – a central, ideally national, repository of data into which brokers and authorized software developers could plug.

One MLS data API governed by one set of commonly agreed upon rules, one approval process and one governing body. Those building on this API would pay royalties that would go to MLSs and brokers. It was, in a sense, like a proto-Upstream.

A lot has happened in the last four years.

This again

Oh, I know: MLS critique is something like debating water fluoridation these days. The issue is old and tired, and other concerns have risen to the forefront.

But I thought I’d revisit this subject in light of all the change happening right now in our industry (plus, honestly, I also read a really stupid MLS critique last week that got me thinking about this again).

Good things that should be recognized have gone down, and there are a few developments that we ought to keep an eye on.

So, here goes. Since 2011:

  • The NAR board of directors required Realtor-affiliated MLSs to provide data to brokers for the purpose of creating AVMs. Some MLSs were strongly against this, many big brokers were for it, and most people didn’t pay attention. But the bottom line is that members can now use MLS data to their benefit in a new way – a win in my book.
  • The NAR board of directors required Realtor-affiliated MLSs to make sold data available via an IDX feed, to permit commingling of MLS data within an IDX display, and mandated adoption of the RESO data dictionary in the not-too-remote future. The net effect of all this is to make it possible for brokers to deliver better user experiences to consumers. A clear win.
  • MLS system software has become notably better. The days of OSX incompatibility and mobile fails are largely behind us. Most MLS systems actually look like software circa 2015. No high-fives here, but things have moved in the right direction.
  • Retsly, a tech company that created an API-based system for accessing MLS data in a compliant manner, launched something quite close to my “socket” idea. In fact, it was simpler and more elegant than what I had described. Sure enough, Zillow bought Retsly in its infancy. That’s not a bad thing, but I would have liked to see the concept grow outside the political distortion field that surrounds Z. Nonetheless, this trail has been blazed. A good thing.
  • Brokers, mostly the large ones, started to get aggressive about voicing – and acting upon – long-held MLS grievances. This is complex and sometimes contradictory stuff, but the net effect has been a greater MLS sensitivity to broker needs. There have always been MLS execs adept at balancing broker and MLS objectives (David Charron at MRIS is a great example), but there were too many that were either tone deaf or simply hostile to broker issues. The broker uprising woke a lot of people up, and that’s a positive development.
  • In the same vein, the announcement of Project Upstream – the planned effort by brokers to aggregate, share and distribute listing and other data before it goes to the MLS – has teed up a new power dynamic. This project is often analyzed as a tech initiative, but strategically, it is about brokers taking a few steps back from the MLS, data in hand, to establish a new point of leverage from which their objectives can be more effectively advanced. This has some MLS folks on edge. And edginess is better than complacency. So even if Upstream flops, it’s opening some new conversations in the MLS world. A good thing.
  • The MLS map still looks like medieval Europe, a patchwork of fortified domains postured defensively. But things are moving. Slowly – glacially even – but MLS leaders like Art Carter and Jim Harrison, who run big MLSs in California, are sharing data with their neighbors and with each other. Brokers and software developers still face unnecessary complexity. But still…

Where does this leave us? I’d say we are in a better place than we were four years ago even in the absence of dramatic change. The airing of broker/MLS dirty laundry was cathartic. We have some evidence that data consolidation may have legs. API-based solutions have shown promise.

My hope for 2019 – more of the same, faster.