MoCoRealEstate

Redfin’s Stats They Don’t Want You to Know

October 2nd, 2011 Categories: Market Updates

Redfin released their Redfin Scouting Report this week which created quite a kerfuffle. With it, anybody on the web can enter an agent’s name in any of the areas they serve and see listings and sales by all agents – not just Redfin agents. By providing detailed agent statistics, Glenn Kelman, Redfin’s CEO, proclaims it “will help you make a better choice about which agent to hire, and hopefully that choice is more often to hire a Redfin agent”.

Well I don’t like it for a number of reasons. My initial objection was that Redfin didn’t have explicit permission to use MLS data in this way. To participate in our MLS (the Metropolitan Regional Information System, or MRIS), competing brokers agree to a common set of rules in order to share listing data to the great benefit of all consumers. That’s how I’m able to provide my visitors with a complete set of all homes for sale in Montgomery County. Redfin broke these rules when they published a view of sales data by all agents for all brokers without explicit permission. MRIS agreed this wasn’t right so they pulled the plug on it. Consumers may not be sympathetic to that argument, but this issue may not be so much about limiting consumer transparency as it is about one broker breaking the rules to serve its own interests.

Here’s what Redfin’s scouting report won’t tell you:  Redfin is not very good at selling homes. That’s what our MLS data says.  If true, why would Redfin – a very smart organization – do such a thing? Either #1 – they aren’t looking at the right metrics when defining success, or #2 – they knew it, but didn’t think they would be called out on it. I believe they were so enamored with this programming creation that they didn’t slow down to really think about what they were doing. Only they know. I’ll show you the unvarnished data so you can decide. But first…

Why I Respect Redfin

Before I start ranting about how I don’t like Redfin, here’s why I like Redfin:

• I’m convinced that they are genuinely motivated to provide good service, save their clients money and still make a profit. They believe that buying and selling real estate is too expensive and it could be more efficient. I think they are absolutely right. I’m not convinced that their business model is the best one to make it happen, but that’s an honest point to be argued. They may well be the future of real estate, or they may not. It’s good they are around to help show another path. Many agents don’t like Redfin because they are threatened by their business model. I am not. The Consumer will pick winners in this industry, and I’m happy to compete.
• Redfin has an excellent consumer website. Their site is easy to use and has lots of accurate and useful data. It is one of the very best and they work hard to make it that way. Based on my anecdotal discussions with buyers at my open houses, I believe it could be the single most popular home search site serving the Metro DC area. Because it is so good, it forces the rest of us to be better. Good for the Consumer, good for our industry.

But They Still Screwed Up

Redin’s brand is built on transparency and consumer empowerment. But I believe that the Scouting Report is anything but transparent. They took real data from MRIS for the Metro DC area, massaged it, and created agent reports. But they didn’t take all of it, and they are completely un-transparent about how they did their math for what they did show. Since they were so gracious to post my detailed stats on their website without my permission, I’m going to return the favor and post theirs. I’m happy to remove it any time at their request.

On 10/1/2011, I pulled all Redfin listing records in the Metro DC area with a list date of 1/1/2010 through 10/1/2011. I pulled everything – listings sold, listings active, listings withdrawn – everything. You can see the raw data here. The summary isn’t very kind.

If I were a seller, it would be important to know that Redfin is not successful helping home sellers nearly 30% of the time. Why isn’t this in the scouting report? I do have some sympathy since selling homes is super hard in this market. Jeannette and I are smart, ethical and work very hard. But we don’t sell all of our homes either. We’re not a broker with a large team, but two full-time agents with an assistant. Since I posted Redfin’s stats for their entire brokerage, let’s all do The Full Monty and look at our details and the summary:

The market is tough and changing, but I’m not happy with failing to sell 13% of our homes. If you work with many sellers, 0% isn’t doable, but 29% definitely isn’t the number to shoot for either.

It Doesn’t Take Much Digging to Find More Bodies

I could stop there, but there’s more to understand about MLS data when making calculations like the ‘average days to sell a home’ and ‘the average number of price changes’. Redfin has 218 listing records for the entire brokerage.  These 218 records represent 204 clients. A typical seller will have one record that goes through several statuses in our MLS. Here’s a simplified summary:

That’s not the only path. If a home hasn’t sold, the original listing can be ‘Withdrawn’ and a new listing created with a new and lower price. The new one will go ‘Active’ then progress to subsequent stages… or be withdrawn again… or sell… or expire.

So when Redfin calculates their metrics, are they looking at the full collection of records, or just the last record that goes to sold? At least in our MLS, that’s important to know. If they were true to their stated values, I would expect them to describe how they do the math. And, of course, to make sure they look at all records – not just the ones that they like.

But What about Redfin’s Buyer Clients?

Redfin easily has more traction helping buyers than sellers in the Metro DC area. On 10/2/2011 (early am by now) I summarized all buyer sales with a sold or contract status after 1/1/2010 and found 1,218 transactions. That’s a lot – six times the number of listings. But how effective is Redfin at successfully closing contracts? Unlike with listings, this is impossible to tell from MLS data. Redfin does post every buyer review online which is a good thing. How many disgruntled buyers don’t provide a review? I have no idea. I would assume they do what they say they do and actively solicit feedback from all clients.

What’s Next

I don’t expect MRIS to allow Redfin to show the Scouting Report for the Metro DC area. As this unfolds, I would expect many to howl that Goliath (the old school, stodgy, and ineffective status quo) beat David (the brave little guy whose singular mission is to help buyers and sellers, e.g. Redfin) because Goliath is only concerned with protecting the same old way of doing things. This would be the wrong narrative. Redfin was sloppy and dishonest by doing what they did.  I think they should pull it from all the markets they serve.

Having said all of that, Redfin did provide a glimpse of what could and probably should be given to the public. Since Redfin doesn’t speak for all brokers, they can never be the one to objectively calculate and publically display everyone’s performance based on MLS data. Hopefully the data included here makes that case. Ideally, I believe MRIS should be the one to facilitate the calculation of metrics like this so it’s consistent and transparent for all brokers. Then, let each broker put their stats out there. If brokers don’t want their performance displayed to consumers, then that’s fine too. Consumers will have the final vote.

UPDATE – October 2, 2011

Refin just pulled their scouting report today, although not for core reasons listed here.  They got a lot of flack from agents calling out bad data.  And, by creating pages for each agent (over 1 million), their agent scouting report was showing up in Google searches when someone would Google an agents name.  It looked pretty spammy, not too classy.  They may try again later.

Posted by Bruce Lemieux

This entry was posted on Sunday, October 2nd, 2011 at 6:25 pm and is filed under Market Updates. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

1. Billy Jalbert said at October 2nd, 2011 at 7:38 pm

   Fantastic response. Trying objectify a business like real estate is tough…but the bottom line is that metrics are usually really good as long as they measure the right things and encourage the right behavior. I just read an article in Bloomberg about a company that placed profits above everything else and didn’t understand why their employees behaved unethically.

   Providing context for those metrics is equally important. You did a great job of that in this response. Sometimes, Sellers really aren’t all that motivated to sell in spite of what the market and their agents tell them. Sometimes their agents just don’t do a good job. Sometimes, Buyers are motivated by more than just price…

   It is a complicated business and real estate professionals do themselves and consumers a disservice by trying to make it sound easier than it is. Please note, I am not saying that you are doing that in any way shape or form here and that this is more just an observation of the industry in which I work. It sounds like you two are a really effective team and I think this is an outstanding response. Thank you.

2. FranklyMLS.com said at October 2nd, 2011 at 9:39 pm

   It has been available to the public for over 3 years in teh DC area on my site over at http://www.franklyMLS.com . Just enter your last name and click on the Sold data link.

   While I must say I am impressed with your research, I would suggest refraining from calling Redfin “bad”, or for that matter any company “bad.” It is a standard Nar code of ethics issue. You can say you are better (in this regard), but generally “bad” is bad.

   And yes I believe they do solicit feedback equally from all transactions, good and bad. They even count people that lost on a house!! Imagine that. A buyer is mad for losing a bidding war and they get a feedback card… I hope they at least wait a week for the steam to dissipate.

   Frank
   Broker FranklyRealty.com

3. Jim Bilbao said at October 2nd, 2011 at 10:28 pm

   Thoughtful, balanced, and well written. Looking down the road when objective agent comparison data exists for Realtor results, I expect we’ll see that agents specialize, and brokers as well. Some will be better at selling high but less fast or reliably. Others will be great at selling low, more fast and more reliably. Likewise, some agents will specialize in clients for whom “money is no object” that want to buy high. Other agents will specialize in clients that buy low. These specialists will market and find consumers differently. They may all get 5 stars for their matching themselves with appropriate consumers, but receive very different performance marks.

4. Bruce Lemieux said at October 3rd, 2011 at 6:39 am

   Frank – I didn’t call Redfin “bad” – far from it. I wrote “Redfin is not very good at selling homes. That’s what our MLS data says.” And then provided the data. And I posted this directly on their blog.

   I’m aware of your site. You’re doing some of the most creative things with a consumer facing MLS site that I’ve seen. A lot of people have tried to crack connecting buyer agents directly to sellers without much luck (since it’s so hard). It will be interesting to see how you do. A success at something like this would be much more disruptive than anything we’ve seen so far. Also, I’ve been amazed that you’ve been able to have so many agents from other brokers provide home reviews – on your site. Impressive.

   Back to agent stats. Your MRIS/IDX license agreement wouldn’t allow you to do this either. I’m guessing that you haven’t been called out on it since they haven’t noticed. I’ve seen this view on your site before, but as I’m there now, I can’t find it for the life of me. It’s not my role to be the MRIS Moral Police. My day job is hard enough, and I don’t have time. Still – if you think that you or Refin are right to display this data just because you have it, then you’re missing the bigger point that I try to make in this post.

5. Daniel Jeung said at October 3rd, 2011 at 8:39 am

   Very interesting and well articulated article. I haven’t checked my local MLS for redfin’s agent stats to see where I stand. If I did, I’d likely be disappointed by my own production levels of late. I guess my response to anyone questioning how productive I’ve been lately would be that regardless of the total sales, it is the consumer’s satisfaction that is the most important metric. I’ve had a difficult time obtaining listings this year, but my buyer numbers have increased dramatically. (Now if only all my short sales would close…).

   I pride myself on taking excellent care of my clients. Over the course of my career, I’ve done 7 sides for one client on her own purchases(non-investment, just lots of moves and all at full commission). I wonder how many consumers would use their redfin agent again if commission was a non-issue?

6. Kevin Lisota said at October 3rd, 2011 at 5:00 pm

   As I prognosticated, the SEO of their competitors has begun, with 163k+ new agent profiles indexed by Google, none of which work with Redfin. None of them consented to have a profile, but Redfin site gets the eyeballs for their name regardless.

   Agents are going to be rudely surprised when they see a redfin.com profile bubble up on their name in Google.

   I just wrote a post about it:

   http://activerain.com/blogsview/2538893/redfin-just-loaded-163-000-agent-profiles-into-google-you-might-be-one-of-them

Leave a Reply

Name (required)

Email (will not be published) (required)

Website