I’ve been traveling more or less nonstop for a couple of weeks, and busy as hell in any event, so I kind of missed the controversy around Redfin’s Scouting Report. I’d suggest heading over to Jay Thompson’s blog to get his thoughts on the product, and the various responses from the real estate industry to it. I don’t have a whole lot to add to the controversy.
But there are at least a few things that most of the commenters are missing on this controversy. I’m far more concerned, actually, about the responses to Scouting Report and what they say about the industry than about the Scouting Report itself.
The first thing that sets off the alarm bells is the criticism that what’s wrong with Scouting Report is that it’s fatally inaccurate. Jay Thompson, for example, writes:
But that won’t fix the data integrity issues that are inherent in MLS data to begin with. The old adage, GIGO – Garbage In, Garbage Out – still rings true. As long as agents keep entering Garbage Into the MLS (and many do, trust me), Redfin won’t be able to do anything but put Garbage Out.
Oh, they’ll have a disclaimer to that effect. We all know how much people pay attention to those.
Of course bad data in the MLS isn’t Redfin’s fault. It is the fault of those entering the data – that being real estate agents… (Emphasis added)
As it happens, Jay is correct in the general: GIGO is the general rule. But… there is something seriously troubling about this angle of attack. I hope Jay walks it back a foot or two.
Let’s leave aside software bugs that Redfin has admitted to (and apparently fixed). Any software can have bugs. That’s not a major issue one way or the other. You know what is a major issue? The revelation that real estate agents apparently are terrible at data entry.
Think of it from a consumer’s perspective. On the one hand, I shouldn’t trust all that agent performance data, because it’s just garbage: real estate agents are absolutely terrible at doing accurate data entry into the MLS. And the MLS with its complex compliance mechanisms can’t ensure data integrity as it comes to performance. So don’t worry about that.
But on the other hand, you should absolutely trust the real estate agent’s website for listings, instead of these third party portals like Trulia and Zillow, because the MLS has the most accurate data on the planet.
The same group of people — real estate agents — enter data into the same system. But one set of data output from that system is Total Garbage and should be disregarded by consumers, but another set of data output from the same system is The Gold Standard of Data Accuracy. Only politicians could hold such thoughts simultaneously in their heads.
This reminds me of the Jeffrey Skilling at Enron debacle, when he claimed that he didn’t know jack diddly squat about financial statements and just relied on the accounting folks. I thought at the time that Harvard Business School, of which Skilling is an alum, and McKinsey & Co, where Skilling was a partner, should have released statements accusing Skilling of lying: “There ain’t no way in hell that any graduate of Harvard Business School and a former partner at McKinsey wouldn’t be an expert on corporate finance.”
MLS’s should consider releasing statements in the near future asserting that the data is 99 44/100% pure and that any inaccuracy is the result of bungling by Redfin engineers. They should stand up on mountaintops and loudly proclaim that between the training, the Code of Ethics, the multi-level compliance mechanisms, and the hard work of millions of professional realtors, the data from a MLS is absolutely, without question, the most accurate thing available.
Comments by people like Matt Cohen are right on the money:
If the MLS (and members) bought into the idea that certain statistics had to be generated and be accurate for display then rules and compliance resources would be put into making sure the basis for those statistics was accurate. Then the data may not be perfect, but it would be better.
No dataset in the history of has ever been or will ever be perfect. But it is critically important that the MLS and the industry as a whole reaffirm that the MLS dataset is the most accurate, the most up to date, the most complete, and the most reliable of all real estate related datasets out there.
The second objection is generally in the category of, “There ain’t no way that Redfin can just publicize all this data without permission, and there ain’t no way that they were granted that permission.”
I’m sure smart lawyers all around the country are hard at work to figure it out. But as I see it, this Scouting Report is a new application, a new use of data that is not contemplated by the license granted to each MLS participant. As someone named Bryan Seiner (I Googled him but couldn’t find the man… odd) writes on the Redfin blog, they’ve looked at this at Redfin and they think they’re in the clear — except in those markets where they are not (and they’ve listed those markets where they won’t display the information). But Mr. Seiner notes that Redfin made the data available to its own agents to use in negotiations and meetings with clients, and points out, “The standard the Department of Justice established in its anti-trust lawsuit against MLSs is that any information we can share between agents and clients in person we can also share electronically with registered customers.”
If you’re angry that Redfin is releasing all this info to the public, then you could probably agitate to get the MLS policies changed to prevent such a thing. But do keep in mind the above about VOW access, what it allows Redfin to do, and what the standard is for making MLS data available.
The important distinction here is that Redfin isn’t a web portal. It isn’t a third party aggregator. It is a brokerage. It is, in the parlance of the MLS industry, a participant entitled to the full rights and privileges of all other participants.
The markets in which Redfin has decided that it does not have the right to display performance data are likely looking at some further work. If Redfin can’t show the data on the VOW, it means that the data can’t be shared with consumers offline either.
Has anyone considered what the possible fallout from something like this could be?
The Franchise IDX Issue from Mid-Year has a bunch of people somewhat… skittish about what could go down in Anaheim. Word is that the rule would be fully repealed, which should assuage the concerns of some of the large brokers around non-participant access to the MLS data.
Now, we have this: participant access to the MLS data, being used in ways that none of the brokers involved ever considered before.
And as the Redfin “morning after” post illustrates, we have a patchwork of rules where some areas do allow the Scouting Report, some do not, and others put certain limitations on it (e.g., two years of data, not three, etc.).
I simply can’t imagine this issue not becoming a topic at NAR. And I can’t imagine this issue not being discussed in executive suites at some of the largest brokerages in the country. I have no idea what, if anything, would come of such topical discussions, of course… but for some reason, that combination there makes me awfully nervous.
Keep an eye out for developments on this front. This could metastasize into something far more.