Initial Thoughts on the Redfin Scouting Report

Some within the real estate vertical are aware that Redfin, a Seattle based real estate brokerage / software / tech company with a presence in 15 or so real estate markets across the country announced the unveiling today of its “Redfin Scouting Report”.

And if you are in the real estate industry and not aware of what Redfin just announced, you should be. It is potentially a game changer of epic proportions.

What is this Scouting Report that Brian Boero of 1000 Watt Consulting called, “the most disruptive online real estate play in years”?

In a nutshell it’s a web based search that returns some sales statistics on real estate agents in the markets that Redfin serves (including the Phoenix metro area). One might even [gasp!] call these “performance statistics”.

Those “stats” are…

For an agent’s listings:

• Currently  active listings
• Average Days on Market
• Median Sales Price
• Range of sale prices
• Average number of price changes
• Number of dual agency transactions

For a buyer’s agent:

• Range of sale prices
• Median Sales Price
• Number of short sales
• Number of REO/Bank Owned sales
• Number of dual agency transactions

All this is contained on a nifty little map and the user can select between an agents listings (sellers) and buyers for either the last 12 months or in the last three years. In addition, links to the properties on Redfin’s site are included.

It is stupefyingly simple to search and see the results. Be advised that the search will only show agents that are within Redfin’s existing market areas, and some of those areas prohibit displaying all or some of this data. For you Phoenix types, all data can be displayed, without even having to register.

OMG and WTF me, someone is throwing out agent sales data for anyone to see?

Yup.

Those not employed in the real estate vertical may well be saying, “So what?” or “Cool!” or “Well, why not?”

Reaction from those who make a living as a real estate agent may just be slightly different… In just a few hours post announcement I’ve see agent and broker reactions that range from, “No problem I’ve got nothing to hide” to “Typical Redfin publicity play” to “I’m glad I have my attorney on speed dial cause these bastards are going down!”

And everything in between.

Here’s my take…

I’m surprised it’s taken someone so long. Whether we like it or not, consumers clamor for this kind of info. Sooner or later someone was bound to offer it up.

It’s not an unheard of product. Diverse Solutions rolled it out at an Inman Conference in 2009 as part of a “code-a-thon” within the conference. The Houston Association of Realtors led by the always progressive Bob Hale did it within their membership — for a week or so until the backlash from agents forced the plug to be pulled.

But now Redfin’s gone off and done it. Putting on public display some performance stats for a boatload of agents across multiple cities and states.

This is going to get interesting…

I’ve long been a fan of transparency, overused as that word tends to be, along with consumers having access to as much data as they need in order to get through the tedious, difficult and expensive process of buying or selling a home. I tend to fall into the “we have nothing to hide” category of agent/broker reactions noted above.

But by golly, the data had best be accurate. Because inaccurate data is not only meaningless, it is dangerous.

And data accuracy is my immediate concern with Redfin’s agent rating product.

They pull data from various Multiple Listing Services (MLS’s). And trust me, if you’ve not had the pleasure of perusing the MLS then you cannot begin to fathom the inaccuracy of that data set. GIGO (Garbage In, Garbage Out) reins supreme within the MLS – and I’m completely confident that’s not just in my MLS.

I’ve seen wrong sales prices entered. Wrong dates. Wrong agents. Wrong brokerages. Wrong type of sale. You name it, it’s been entered wrong. If you are pulling a data set and then doing calculations like median sale price, days on market, price range etc. well then you need accurate data to begin with.

Here is what I found:

Using the shiny new Redfin Scouting Report, I looked up about 50 or so Phoenix agents. A large number of them (I’m estimating 20%) weren’t even associated with the correct (or any) brokerage.

When one sees that something as fundamental as the brokerage is amiss, one should be wondering how accurate any of the other data is…

Days on Market seemed WAY off everywhere. Redfin later admitted there was a bug in that, and squashed it swiftly.

When one sees a fundamental calculation with a glaring error on release, one should be wondering how accurate any of the other data is…

There appears to be little to no recognition of agent teams, of which there must be tens of thousands in the Redfin market areas. The handling of data on agent teams is a notoriously difficult problem in this space. The “team leader” may well get all the “credit” for everything his or her team members do. Of course this means the team members get no “credit” for anything they do. I’ll be honest, short of fundamentally reworking the MLS data structure, I don’t know how this can be fixed.

So the data isn’t perfect, so what?

Well, this is a tool designed to help consumers select an agent. Leaving aside some huge arguments that “performance data” is far from the only thing that matters in agent selection, what we now have is a real estate brokerage that is reporting publicly on agents from competing brokerages – and using data that is flawed on some unknown level for those reports.

This my friends, is not so good.

I don’t believe that Redfin’s disclaimers like, “This scouting report includes information from ARMLS for transactions since Jan 1, 2008 and is accurate to the data feed Redfin receives from the Multiple Listing Services” means a damn thing to the average real estate consumer. They may just be inclined to think “ARMLS feeds accurate data” (assuming they have a clue what ARMLS even is…).

I like Glenn Kelman (the CEO of Redfin). I’ve had the pleasure of meeting Glenn and interacting with him. He’s a stand up guy and an absolute master of public relations. I’m glad he’s got a boatload of venture capital to create cool tools like this, and to hire very good attorneys. What will happen as soon as one agent finds out they lost out on a potential client because the scouting report showed flawed data? Only time will tell, but it will happen. And personally, I don’t think blaming the MLS for being a crappy data source is going to hold much water.

But maybe that’s just me.

What if the data issues can be resolved?

Redfin certainly isn’t going to kick back, put their feet on the table and call this project done. They’re a tech company and they constantly measure, tweak and improve things. Heck, they are one of the few out there with the smarts and capital (God bless those VC angels!) to work through these issues. And I suspect that they will do just that.

In fact, I hope they do.

I’ve got absolutely nothing against Redfin. Never have. Even during the years it was cool to hate on Redfin I wrote about them on occasion, but other than for the most asinine real estate blog post ever written, I’ve never had a bad thing to say about Redfin. I welcome their innovation and different brokerage model.

I’m a big proponent of agent reviews and ratings. I even set up our own cheesy site to help facilitate collecting client reviews of our agents.

Ratings and reviews are here to stay. Consumers demand them, someone has to deliver them. I wish Realtor Associations had taken the initiative to do what Redfin is trying to do here. To me it makes more sense for an Association to do this, not a brokerage. I think Redfin is going to get a lot of backlash from less progressive brokers and agents over this, and that comes at a time when Redfin is far less maligned then they were in the early days.

Assuming Redfin can resolve the data issues, and assuming they add some education and comparative stats to their search results (is 47 Days on Market good? Above average? Below?) then they are on to a winning thing here.

Is this the “most disruptive online real estate play  in years”?

Quite possibly…

And hey, I like disruption.

I think this industry can use a little of that.

What do YOU think?

I have more to say on this subject. Stay tuned for Part 2: Why Refin’s metrics might be wrong, and what would work better.

Originally posted by Jay Thompson on Phoenix Real Estate Guy. Definitely visit his site for more great content and an excellent example of an effective web presence.