Earlier this week in a post comparing real estate sites Trulia and Zillow, I suggested that the most important success factor for these sites is how comprehensive they are. The more listings the better because home buyers want to go to one place to find every home on the market. They want a single dashboard from which they can filter down the choices.
But just how comprehensive are these sites, and how accurate are their listings? Trulia offers 3.5 million listings nationwide, and Zillow has 3.1 million. But people look for houses in local markets, not nationwide. What matters is comprehensiveness withing local market
A few hours after I put up that post, I heard from Roost, a competitor to both Trulia and Zillow (with only 1.4 million listings in the markets it covers) that differentiates itself by getting its data directly from the same Multiple Listing Services (MLSs) that real estate brokers use. They just happened to have study in their back pocket (which they commissioned and paid for) that compares the “accuracy” of search results in three cities (Dallas, Miami, and San Diego) across different real-estate search services (Roost, Zillow, Trulia, Yahoo, and Google). The study, which was done by real-estate industry consultants the WAV Group, defines accuracy as the percentage of listing results that match listings the MLS for that city.
The results are in the chart above and, not surprisingly, Roost comes out looking great. For each city, it returns between 95 and 99 percent of the listings in the MLS. Trulia’s accuracy in the study ranges between a pitiful 9 percent for San Diego to 61 percent for Miami. (Zillow generally does worse across the board, with its accuracy ranging between 12 percent and 36 percent across the three cities).
Trulia disputes these results. Heather Fernandez, vice president of marketing, says:
The data looks very questionable, and not in line with our internal coverage data. Our data shows that we have roughly 70% coverage in most major metros.
And indeed, if you do a search for homes for sale in San Diego on Trulia, you get 4,395 results, compared to 6,036 on Roost. That’s 73 percent. (Zillow claims 7,661 listings in the San Diego city limits). Even if half of them are stale listings or not accurate in some other way, it’s hard to get to the 9 percent that the Roost-financed study claims. That’s because for some reason, the WAV study only compared homes in each city with exactly 3 bedrooms and 2 baths, within a $50,000 price range.
That methodology seems random and flawed to me. Would Trulia’s accuracy be greater if the study had looked at homes in San Diego that cost $400,000 to $450,000 instead of $300,000 to $350,000? Comprehensiveness would have been a virtue in the study’s methodology as well as in what it was trying to measure.
Still 70 percent accuracy is not that great, and it doesn’t seem like Zillow is any better. If the MLS in any given city is the benchmark, both have a lot of work to do. And Trulia, for one, is striking deals with different MLSs to incorporate their data. But it only has 14 so far, out of about 900 nationwide. MLS-based sites like Roost and Redfin may have more listings in the markets they serve, but they don’t serve every market yet. For instance, in San Diego, Redfin tracks 6,300 homes for sale, better even than Roost. Yet neither has any listings in New York, and Redfin only has 473,000 listings total.
Yet Redfin CEO Glenn Kelman also balked at my earlier suggestion that either Trulia or Zillow are even close to comprehensive within any given market. In an e-mail to me, he said:
What got me was that almost any real estate site has more homes for sale than Trulia or Zillow.
Frenandez doesn’t think that having the most listings matters. She responds:
Listings are commoditized –there are dozens of sites that offer basic listing information in every city across the country. It’s not a competitive advantage for an Internet company.
What is more important, she says, is the filtering the site allows home buyers to do to help them make an informed decision. I’d say it’s both. Those filtering tools (heat maps, sales comps, local school info) are also becoming commodities. You want to cast your net as wide as possible before you filter down so you don’t miss out on that one house that fits all of your criteria.
See all
Post
Aaarrrggg…the real estate industry is so anti-competitive, NAR in particular. They’ve used data access as a weapon to protect their monopoly for too long. This data fragmentation and poor coverage is the result.
We need more open standards, innovation, and competition in the real estate industry. Make Realtors earn their keep by providing services for a flat fee, rather than this percentage nonsense, which is driven and protected by the stranglehold that NAR has on the industry.
Imagine online travel reservation 15 years back, it seems like piece of cake from the moment we put the dates on expedia or orbitz.
Why can’t the real estate industry try to jointly develop a comprehensive business management software, a centralized repository of housing information and then companies can build tools and services over. Well might take 10 years - but will make the job of our kids finding houses easier.
http://mydatingtree.blogspot.com/
Real Estate Transaction Standard (RETS) is trying to create an open standard - http://www.rets.org - I hope they do.
From reading your post about the Roost study it definitely seems flawed. If they were after determining how comprehensive the listings on both Trulia and Zillows sites you would think they would conduct the same type of comprehensive type study. I know that it was probably cost prohibited but it makes their claims look a little silly.
I think that places like Trulia, Zillow, Roost and many others are best used by the public as good starting points in their real estate search.
Use the one you like the best. There is no one best or one size fits all search tool out there.
These sites that are basically doing comps, need to aggregate as many sales stats as possible. I would never rely soley on any of these sites for buying or selling a house, only as a reference. “Doing homework”
Is the complete study available?
There are plenty of sites to sell your own home-FSBO.
Somewhat related: Most sellers need the professional realtor to sell as the paperwork can be overwhelming to most people. It would be interesting to see how many people are really satisfied with their last real estate agent - transaction. There is plenty of room for improvement. There are really no standards. The MLS sites are all different.
@James- I think the RETS standard would be a great thing…it would help a lot of internet companies as well as traditional brokerages with a presence in more than one state and MLS.
That’s not a study, its PR. The bigger question is why all these companies are trying to emulate the MLS. The real opportunity lies in breaking it.
WAV group is funded by MLS and release study that shows MLS listing data is more accurate than Zillow’s ( Zillow is not a paying customer of the WAV group ) - Wow! I didn’t see that one coming.
http://wavgroup.com/Home/Customers
Agreeing with Shuki: The sites are a great starting point and have good potential to yield useful results, but talking to someone on the ground, so to speak still might be the best end game at this point in time.
I’ve generally used Yahoo and Trulia, if not the MLS. The others just don’t have listings for places I’m looking at. Yeah, it makes sense to start with the larger markets if you have to limit yourself, but isn’t it easy to assume that there are still more people in total looking for real estate in all of the combined smaller American markets than total looking in the top five or ten?
I never understood why any online company would tackle real estate from a national comprehension point of view. As a consumer, why do I care that you have 80% of listings in the US? I only want to buy a house in the town I live in. Why not start with giving full access to listings one city at a time, then expand to other markets?
First off, our intention wasn’t to say that Trulia and Zillow are bad real estate sites. Both are awesome sites, with heat maps, real estate wikis, mortgage tools, Zestimates and plenty of other resources for real estate shoppers. We just took exception to the idea that they show nearly all the homes for sale, since this is where we try to shine.
One reason Redfin chose the business model that it did, to be an E-Trade for real estate, is because it allowed us to access the listing data that other Realtors access: every broker’s listing, all the photos of each listing, and complete property details.
We don’t think you can build the best real estate search site without this data. At the same time, we are not sure we can build a better business than Trulia or Zillow when we actually have to do the work of representing customers buying and selling homes. You have to work your fingers to the bone providing fanatical customer service. And the margins are worse. In fact, our margins are terrifying. It’s a trade-off.
As Ryan Waggoner points out, we have worried that accessing this data would preclude us from publishing other types of information, such as how long the home has been for sale or what other people think about its price. We used to rant and rave about these limits, getting fined, and generally making trouble.
But Ryan isn’t completely right. Recently the rules have liberalized: we now show all the homes for sale, including ones not represented by brokers (foreclosures, for-sale-by-owner). We show how long the property has been on the market, pictures of the home from other sources, Zestimates from Zillow, which is our partner.
Other Realtors have given us free access to listing data even though we charge a radically reduced fee. Earlier this summer, the National Association of Realtors struck an agreement with the DoJ to provide listing data regardless of a broker’s pricing or business model.
The rules still have a long ways to go, but the fundamental limit on media-driven sites isn’t just these rules. Media-driven sites are, to varying degrees, pro-Realtor. Such sites run ads for Realtors, who would never place those ads if those sites showed all the information that the Realtors show. And a site running Realtor ads almost never provides data that disparages a listing because it would also upset the Realtor.
That Redfin – which has said terrible things about the real estate industry on 60 Minutes — has the freedom to access all the listings, and to provide different perspectives on each one, surely says something how much the real estate industry changed.
And please don’t call me a sell-out just because I’m talking about ways real estate has gotten better. Redfin still wants to be an agent of change in the industry and in every possible way, we will always be an advocate for the consumer. I just don’t think it’s as simple as saying Web good, Realtors bad, particularly when many websites are limited by their advertisers as much as Realtor rules.
Written in haste, at absurd length, but sincerely,
Glenn
Glenn Kelman
CEO, Redfin
Hey Erick - thanks for pointing out the silly methodology of that study and letting us respond. In fact, I was thinking about commissioning a study saying that I’m a size 2 supermodel.
But seriously, we agree that comprehensive listings are important and we are working with hundreds of MLSs across the country to increase coverage (but even they don’t necessarily have 100%). The point I wanted to make is that comprehensiveness alone doesn’t make you a winner in the Internet space.
As a side note - due to our relationship with NY brokers and the Real Estate Board of New York (REBNY), we have the most comprehensive NY search of any national site.
I’m looking for a place in NYC and I’d give you 3/5 for listing coverage in NYC, even if you have a relationship with the real estate board. In my experience StreetEasy has better coverage. I find many listings on there that you just do not have.
@heather “we have the most comprehensive NY search of any national site” is the most disclaimed statement by a person thinking of commissioning a study claiming themselves to be size 2 supermodels.
Seriously.
Why would I want to search NYC real estate on a national site if there is a better local alternative?
@ sebastian - we know from our internal data that people who use Trulia use other real estate sites too. You work at a street easy, right? I’m guessing you would see the same thing. So the quick answer to your question, is that they might use both.
NYC is consistently the #1 or #2 city on Trulia. From user feedback, they use it for listings, comps, searching on the go, Trulia Voices community and cool apps that we continue to roll out.
Now from advertiser perspective, that means that people from across the US and globally are searching on Trulia for NYC properties (which happens to be a happening global spot). If I’m a brokerage/builder etc - that makes Trulia a compelling advertising platform.
Heather, silly is mixing in RealtyTrac data with actual listings for sale and then claiming to be comprehensive. Most of your RT data is pre-foreclosure stuff that IS NOT listed for sale.
Try this in NYC, where there is no MLS! StreetEasy has the best set of listings here, but even then, the entire market is just a mess.
It isn’t surprising that Realtor.com was not part of this comparison study / PR effort! If you search for San Diego houses on beta.realtor.com, you get 6,331 listings.
Its the most comprehensive national and local site. No questions there.
The new REALTOR.com does look great.
Realtor.com uses the same MLS IDX data feed as Roost and Broker websites. Hence their data performs the same
@Ajay- I have to say I am impressed with how far along you guys have come. The new beta looks great.
ziprealty.com ain’t bad either. I am in the process of closing on a house in Miami right now, 144k, was sold for 400k in March 2007!
Zillow added listings via bulk feed late last year and although the program is new it has seen fantastic adoption.
If you’d looked into this issue, you should have also learned that not all homes for sale are in listed the MLS’. To define “all listings” that way is disingenuous (at best and monopolistic at worst.) Stats vary but by most accounts, the MLS contains a maximum of 80% of the homes that are listed for sale. The other 20% are FSBO’s and pocket listings that aren’t on the market long. Many of those listing are on Zillow, do not appear in the MLS but do appear on sites like Redfin - thanks to data feeds from Zillow.
As others have remarked, this data is ridiculously questionable. It looks like you’ve been played on this one, Erik. That’s twice in one week. In future, please contact Zillow for comment.
The data is flawed, as I point out. But it does raise some interesting questions.
When I want to be played by Zillow, David, I’ll be sure to contact you for the data that makes Zillow look good
“…As others have remarked, this data is ridiculously questionable.”
Hey that’s ME!
Yes it is true, 20% of homes listed are FSBO’s until they don’t sell — then they are listed by Realtors who get them sold.
I got on this site because I want to know where and how zillow gets their information. I would love to sell you my house, no questions asked, for what zillow says it is worth!!! Zillow says it’s worth $13,500 more than what I KNOW it would sell for.
So how do you get your information?
correction: “not all homes for sale are listed IN the MLS’”
I thought the for sale by owner market was much larger than the agent market. Most of the listings on the MLS are from agents.
Does any of this actually matter if nobody is buying homes?
Any site that cooperates and synchs with the MLS will be leaps and bounds better than a site that doesn’t.
Zillow is a great resource for historical data, but their listings must be manually input by each agent. If there is a price or status change, that agent must remember to login and update, or the data will be incorrect.
Which do you think works best?
The fact that we’re even having this discussion in 2008 is an unfortunate artifact of decades of control-the-data-and-control-the-customer mindset of the real estate industry.
Companies wanting to display listings — whether as a broker (a la Redfin) or as a media play (a la Trulia and Zillow) or as a hybrid model (a la Roost) — have to make a tough decision up-front:
Do you get your data from:
a) The MLS?
b) Brokers
c) Consumers?
a) is always the most comprehensive, so if your goal is comprehensiveness (Redfin) you pretty much have no choice but to go this route. (Redfin, as a brokerage, has a gazillion other reasons in addition to comprehensiveness why they would have to do this.) The downsides, however, are legion: The US has a patchwork of some 900 data fiefdoms (the local MLS’s), each of which has different data standards and different display rules. In City X you can display solds, but 50 miles away you can’t. In City X you MUST show the type of siding the house has, but NOT the type of roof. Fortunately, thanks to the superb air cover provided by NAR, Redfin has been able to push the boundaries of these insane rules and has probably been the cause of some rule changes (though I’m sure Glenn Kelman wouldn’t be able to publicly comment on that.)
b) There may be 900 MLS’s, but there a gazillion brokerages, so you immediately have scale problems. Thankfully (at least wrt solving this problem) if you can sign up the largest 15-20 nationwide brokers you’ve probably got 70% or 80% of the inventory. But unless you’ve got a sales force the size of a small city, you’ll never get too much above that; it’s cost-prohibitive to ink agreements with every mom-and-pop brokerage around the country. On the (very big) plus side, you don’t have conflicting MLS rules to manage, plus (I suspect) you don’t have to pay the brokerages for access to data; they’re happy to provide it in exchange for broader syndication of their listings.
c) Very Web 2.0, and potentially scalable, but only if the vast majority of the US were Web 2.0-savvy, which of course they’re not. Plus, I can’t imagine consumers being happy with the notion that, despite having hired an agent, they have to upload information about their home to a number of sites.
While the numbers in this study do appear a bit skewed against Trulia and Zillow, a quick and dirty informal study will show that, by definition, Roost has more comprehensive coverage (in the cities they’re in) than Trulia and Zillow, though certainly not by the margin indicated.
On a further note, I can attest that Heather Fernandez of Trulia is indeed a Size 2 supermodel.
One of the major problems is that most MLS’s technology is out of date / just not there and is very difficult to get that content and when you do it may not be too accurate.
At the Inman Connect show in San Francisco this was the exact issue that came up which is a very touchy subject. The whole basis of the debate was that Realtors want to get the content/ data out to the masses and others have a problem with giving up control. There clearly needs to be a happy medium but, the correct solution is anyone’s guess.
Which search engine do you use: Cuil or Google? Cuil hit the market with much fanfare trumpeting the number of pages they index. Similarly, in real estate search, some sites lay stake to being “comprehensive” because they have the most listings nationally. hmm… I’m searching for a home in the Peninsula… is it relevant how many homes a site might have for Buffalo, NY? As a homeseeker, my priorities are:
1. Number of listings in my search area.
2. Accuracy of listing information (very frustrating to spend time evaluating a listing if it is already off the market)
3. A user-friendly, intuitive search
Heather@trulia says listings are “commoditized.” Respectfully, consumers think listings are commoditized. then why just “70% coverage?” If I’m searching in an area with 100 potential houses, I never see 30 of them. These might be the best deals too since lots of real estate search sites don’t list these houses. Listings are a commodity? That’s a farce!
I would be less paranoid except I’m trying to purchase a home—perhaps my largest and most important financial decision. Is it safe to do a partial search to find a perfect home?
Correction on my above comments:
As David G. from Zillow above points out, most MLS’s don’t display FSBO’s (For Sale By Owner properties), so it is unfair to define “comprehensiveness” as “% of MLS listings displayed.”
A better definition is perhaps “% of ALL listings displayed.”
If indeed 20% of homes on the market at any time are FSBO’s (that sounds a bit high to me, but David G, though not a size 2 supermodel, is usually spot-on with his data), then in order to win the “comprehensiveness” race (if it’s a race worth winning) a site like Trulia and Zillow would need to have 100% of the FSBO’s in order to make up for only have 80%-ish of the MLS listings.
Redfin would then be the clear leader, as they should have 100% of the MLS listings (in the markets they cover) plus whatever percentage of the FSBO market they get from Zillow.
Dan Sullivan -
The vast majority of the 3M listings on Zillow.com are not entered by hand into the site. They are fed via XML feed from brokers, MLS’ and listings aggregators.
And don’t forget that no matter where a listing comes form, someone entered it by hand somewhere along the line.
Most of the real estate sites mentioned have some cool features but without all of listings listed from the MLS I would say their value is limited.
If you want to see what a quality real estate site looks look visit
http://www.har.com This is what all cities should have.
Great Comment - our firm was invited to ask consumers questions about why they value HAR.com so much over other real estate web sites.
They said (Choose all that apply)
59.1% - Has all Houston Listings
54.1% - It is easier to use
51.1% - No salesperson will call me
44.9% - Information is up to date
22.1% - Can contact listing agent
14.9% - Other
Furthermore we asked “Where else do you get information about real estate in the Houston Area (Choose all that apply)
21.6% - Newspaper
20.6% - Realtor.com
15.4% - Real Estate agent websites
15.1% - Real estate company websites
14.1% - Other
13.4% - Zillow
12.9% - Real Estate Magazines
10.7% - Craigslist
6.2% - Homes.com
3.7% - Trulia
Thank you Erick for the post. For starters, we’re very confident in the methodology of this study – and stand by it. Victor from the WAV Group will respond specifically on the methodology to address related comments.
http://www.Roost.com does one thing well – provide consumers fast access to comprehensive homes for sale data. Consumers in this space are real people looking to make the biggest investments of their lives. It’s not just about eyeballs and ad dollars, it’s about helping consumers find their dream home by providing information they can trust.
There are a bunch of great websites in the real estate space that are doing wonderful things for their consumers. Value estimates, community information and facilitating dialogue with industry experts are all valuable features. We’re not saying Roost is the only site to use online. We’re saying that if you’re ready to get fast access to comprehensive homes for sale data Roost is the site for you.
Erick and others in comments nailed it. Real estate is local. If I’m searching for a home in Denver, I don’t care if the site has 3 million listings nationally – or a bunch of homes for sale in San Diego. What I care about is the site I’m on has all the homes for sale in Denver. Yes, no site will ever really have all the homes for sale. But, considering that the MLS is the standard for listing homes for sale, sites like Roost will continue to have the most comprehensive data online.
As of last week, Google isn’t even a player in the market anymore. They took out almost all of their filtering tools, so you can’t filter based on home size, lot size, or many other filtering options - that they had two weeks ago.
One thing I noticed on zillow is property valuations differ from sales prices so even on a recent sale the value can be higher than what it actually sold for. When foreclosures have become such a large part of the market they are not really nonmarket anymore.
Good post and a good topic. I recently looked into refinancing my 1BR/BA condo in Downtown San Diego. The SD Credit Union has a independent appraiser they use. This guy came in took pictures and measurements. His report valued my place at $56K below what Zillow lists my place. That’s a pretty big swing on a small place. I have no interest in selling, but it did kind of screw up my refinancing process. On the plus side, the City of SD owes me some overcharges in property taxes….
Good post and great comments. Jeff (TheXBroker) hit upon this same issue a couple of months ago:
http://thexbroker.com/2008/03/.....ndustries/
Page down and read the ‘Where is There?’ section…
Also:
http://thexbroker.com/2008/04/.....-listings/
Nice to see TechCrunch catch up
What people seem to have forgotten in the competitive world of real estate websites is that MLS websites and Remax etc are not the only vendors for accurate MLS access. Locally targetted real estate websites by brokerages and individual agents still take up the majority of SERP results(although trulia and zillow are encroaching here), and the majority of these sites utilize MLS Access. Either a link into the MLS provided search system, or a MLS system with data fed to it from the MLS itself. Given that the MLS obviously has 100% accuracy, the question is how Trulia and Zillow propose to overcome the more locally relevant real estate agent sites when they are Geo-located, AND have more accurate results.
As someone who was in this game a while back, I know that MLS listings are by no means 100% accurate. So, that is a problem out of the gate. Realtors don’t key in the data properly, are the base data is wrong. A good service should supplement MLS data to provide a more accurate picture of the property, which is the core piece. From there, it is deal points, and market nuances, coupled with the actual listing that bring it all together for the buyer. I agree that it is the listings in my area that are meaningful to me. That said, a core metric is total listings for companies that are going national, and that is the grail for many…
Every site is has different comprehensiveness, but that is only one thing that may be important to home buying consumers. Although we try to get the most comprehensive number of listings (from the MLS IDX feeds directly), we also pay a lot of attention to how easy is it for consumers to get to the information, and other pieces of information that they may want to look at when they are searching for homes.
We have found that many consumers like our site, and I think what will develop in the internet real estate space is that consumers will have their personal preferences and will go to the sites they like, not that there is anything wrong with the other ones.
If you have not check out our site you should do that:
http://www.movoto.com
Thanks for covering this story Erick. It would be great to get some feedback from Michael Arrington, your founder and co-editor on this topic. As he knows from his experience with his company Edgeio which was a competitor to Trulia, Zillow and others, aggregating listings is a difficult and challenging endeavor for any technology company.
Roost commissioned our company to compare the listing results that a typical consumer might get when in searching for a home in San Diego, Dallas or Miami. In order to do a comparison, we needed a benchmark. We choose the area MLS as it is generally accepted as the most accurate and complete source of single family homes for sale by area real estate brokers. Although it would have been nice to compare every listing in each of these cities, that was beyond the scope of work. Our methodology involved doing an identical search on a local broker website, Roost, Trulia, Zillow, and Google at the same time on the same day. This analysis was designed to begin to help the industry understand the relative data comprehensiveness of third party sites versus those that were “fed” with MLS-provided listing information.
Dallas Search – Single Family Home for Resale (no FSBO or Foreclosure) $250,000 - $300,000 – 3 bedroom, 2 bath.
Miami Search – Single Family Home for Resale (no FSBO or Foreclosure) $450,000 to $500,000 – 3 bedroom, 2 bath.
San Diego Search - Single Family Home for Resale (no FSBO or Foreclosure) $300,000 to $350,000 – 3 bedroom, 2 bath.
While this analysis should be in no way considered a comprehensive study of data accuracy, it is a small step toward revealing the differences between third party sites and MLS-fueled sites. Our analysis revealed that sites with data supplied by MLS information are more up to date. Sites that are not fueled by MLS information suffer from systemic inaccuracies.
As a consumer this can lead to a frustrating experience. Sometimes a consumer will find a home they really like only to find the home was sold several months ago. This experience can cause frustration and can have negative implications for the listing agent and the brand of the broker.
Many third party listing aggregators are now courting MLSs to minimize the problems created by relying on a variety of sources to provide listing information.
WAV Group stands ready to perform additional analyses to continue to help all of us how to continue to improve the consumer property search process.
I still think Zillow is best.
http://www.cityquery.com
Shouldn’t the title of the story be “How Comprehensive Are Listings On Real Estate Sites?” Correct me if I’m wrong but you’re not comparing the matching accuracy of each of these sites but just looking how many results do they return for certain queries?
Let’s take a quick look at San Diego …
- In Q2, 31% of the homes sold in San Diego were foreclosures yet this study ignores foreclosure listings. Could that be because there are not many REO properties in the ROOST / MLS inventory?
- The average home value is $420K yet this study cherry-picked 3-bed 2-ba homes under $350K. Let’s face, the sample used in this study is hardly representative of the real estate market and clearly contrived to produce this outcome.
- ROOST currently lists 6118 listings in San Diego.
- Zillow currently lists 7661 for sale listings in San Diego. 104 of those are FSBO but many more are Foreclosures.
- Zillow also lists 732 Make Me Move listings in San Diego. Those listings can not be found on any of the sites mentioned in this post.
The first sentence on Roost’s website says “Every home for sale …” Clearly that is a lie and this study is an ill-conceived and biased attempt at creating BUZZ around Roost’s false claims.
Drew -
Good top hear that Roost is committed to being a comprehensive listings resource. Please get hold of me regarding publishing Zillow’s FSBO inventory on ROOST. [davidg AT z DOTCOM]
Geographic strengths and weaknesses are a given for all of the early-stage companies mentioned in this post. IDX sites are limited to the areas where they are an MLS participant (like Redfin), or have a lead generation relationship with one (like Roost). Aggregators are limited by their relationships - which may be with a set of the brokers or franchisors who operate in an MLS, the MLS itself, or other aggregators.
And therein lies the problem for anyone using a big listing count number to promote the idea of comprehensiveness to the consumer. Other than the IDX sites named in this post, each of the sites named may be getting the same listing from i) the broker, ii) the agent, iii) the franchisor, iv) the owner, and v) another aggregator. Compounding this problem is that many aggregators - typically those who take listings from agents for advertising purposes - themselves may have several copies of the same listing, in many cases from several agents all purporting to be the listing agent.
So the real measure of completeness, in addition to geographic footprint, is *deduplicated* listing count. It doesn’t do anyone any good to look at a set of search results with five or six photos of the same exact home. It’s not clear to me whether any of the numbers being thrown around here are deduplicated numbers, or whether they account for any measure of accuracy or timeliness.
BTW, the FSBO issue is mostly a red herring. While it is true that 10-15% of listings at any moment in time have been offered by owners, the vast majority end up selling through the MLS.
Hello Marty.
“FSBO issue is mostly a red herring. While it is true that 10-15%”.
Not quite. If you look at NAR annual reports they state that 12% are successfully “Sold” via FSBO, so that would mean that there is a significantly larger amount that 10-15% that are offered for sale at any one time.
Although, I agree the duplicate issue on the same site is a pain… on different sites the same listing can be presented in many different ways which can actually help consumers. Your Cyberhomes is probably in the best position to provide the most complete database considering your access to data
Great topic Erick. Comprehensiveness is extremely important as both David G & Kevin point out, FSBO’s and pocket listings make up a major part of the available inventory and for a homebuyer seeing only 70% of available listings is unacceptable for the largest decision of a their lives.
This explains why the average home buyer searches multiple sites and different sources as pointed out by Mr. Lunds report here:
“21.6% - Newspaper
20.6% - Realtor.com
15.4% - Real Estate agent websites
15.1% - Real estate company websites
14.1% - Other
13.4% - Zillow
12.9% - Real Estate Magazines
10.7% - Craigslist
6.2% - Homes.com
3.7% - Trulia”
…and this fact is also supported by NAR’s annual Internet Home Buyer Reports also. This is why we our beta site http://www.retrove.com is taking a complete different approach to real estate search. We are not an individual listing aggregator trying to be a destination site but a true vertical real estate search engine helping consumers easily find all possible listing sources in their area in the quickest manner possible.
We are the only site that allows you to perform a single location search and get results across all types of sites included all of those all of the sources indicated above. This eliminates that need to find different domains and perform the same location search across multiple sites which saves a ton of time for consumers. We index the results across all sites including individual MLS, agent IDX sites, FSBO sites, newspapers etc. which guarantees the most complete listing coverage and access. Currently we have over 12 million individual search results in the index from 1000’s of local real estate sites. Additionally with every search we provide localized community tab that provides the best sources to research schools and other important information simultaneously.
So instead of trying to reinvent what consumers are already doing, we are simply making the search process more efficient for users. In the search process the home buyer may see the same listing across multiple sites but sometimes reveals different descriptions, photos and prices which helps homebuyers do their homework when it comes to specific listings.
The largest issue affecting most of the aforementioned sites is the politics of aggregating and displaying the individual MLS listing data from over 900 individual MLS’s and FSBO data and the de-duplication of dynamically changing data is a near impossible task. They are all great sites and each bring great advances in real estate search but none can truly be comprehensive in scope when displaying the individual listings.
what a terrible idea Jessie. I can’t possibly imagine this even remotely being cool. Besides the fact that IT DOESN”T work when searching for homes in 98005 (i get query strings that point to no results), anyone searching for a home in a MLS will get ALL results from that MLS. Duh!
Seriously, I can’t believe you wasted time building this out.
Hello Mike..
Well, that thats odd? I just a search for Bellevue WA down to your zip 98005. I got 193 different websites with possible listings.
In that result set, I clicked on each of this links and got the following sample which worked fine?
bank of america reo site - 0 properties no bank reo’s there.
Realtor.com w/ 66 listings
Windemere.com w/ 142
Remax / realtor w/ 73
realestate.nwsource.com w/ 2 listings
estately - 1072 listings
seattlehome - 32 listings
redfin - 164 listings
servingseattle (john l scott) w/ 88 listings
craigslist - 0 listings
trulia - 119 lisitngs
zillow - 1076 listings
So maybe you don’t think it’s “cool” but it does work and it saves lots of time… as the above would take you 20 minutes to type each url and location.
Any other feedback as to what we could do to improve would be appreciated. If you note above, the listings from john l scott, redfin and windemere are mls listings directly via an idx feeds.
A lot of great comments here. I tend to agree with most of the industry regulars (hi Drew, Jessie, Marty, Glenn, etc.) about the sad state of listings in the real estate industry. But, that will improve. It’s just a matter of time.
Marty does make a great point about de-duped listings. But there are so many points to consider here that a comment won’t get to all of them.
I do think, however, that “accuracy” is really a bad term to use here; “comprehensiveness” is fine, but “accuracy” is not. The reason is that the state of listings in the industry is really quite horrible if you think of actual accuracy — i.e., is the listing actually describing the property in full? Are the lot dimensions actually correct? Are the features actually the way they are listed, or is there a ton of marketing fluff in there?
When the listing says “waterfront views”, but the only way to see the water is from a tiny window in the attic… is that accurate? When a listing says 4BR, does it make clear that one of the bedrooms is a converted closet?
It’s probably too much to ask of human beings to be completely forthright about the actual state of a property on a website advertisement (which is what a listing is). But because I do think there is room in the industry for a brokerage that emphasizes such frank honesty, I believe it’s important to preserve the term “accuracy” for that quality: closeness to the truth.
-rsh
I don’t have a dog in the Zoost-Rillow-Wulia fight, but as a homeowner and a citizen I’m glad that someone is beginning to shine a light on the be NAR mafia, its monopolization of market data and other anticompetitive practices, and the $90m or so they shovel into Congressional campaign coffers each year.
Breaking up the 6% mafia and giving the public access to this data should be a national campaign issue.
There is utterly no reason that the process of selling one’s home should result in brokerage fees of more than maybe $1,000, tops. Were this industry forced to open up the transaction data to consumers, and were municipalities to adhere to national data standards, residential RE transaction costs would tumble overnight, and save American homesellers at least $40b per year.
Which is a far better and more effective stimulus package for homeowners than the gimmicks our representatives are toying with now.
Red and Blue voters, unite! Smash the NAR. Write your congressman.
Rob Hahn - “the sad state of listings in the real estate industry. But, that will improve. It’s just a matter of time.”
You’re an optimist. This is not a technology issue. It’s 100% about politics.
There is corruption at work here. We have a de facto system of fixed rate RE commissions, enforced by local cartels of realtors who have bought off the US political class and who ruthlessly punish any broker or intermediary who dares to lower the transaction costs and share accurate transaction data, ie the MLS data with consumers.
The solution will come from DOJ and with Congress, not from entrepreneurs.
Dude,
The major brokers ALL use the same stuff. They may refine it, but its the same idx feed (nwmls). The best thing you can do for anyone would be to go to an independent brokers idx feed, like seattleshomes.com and search. There you will get an exact record of all the NWMLS homes for sale in the region. Or, you can search nwrealestate.com search, which IS the nwmls MLS itself.
I looked through your js, looks like you just search google web and google base and organize your listings via a ranking system.
Nice, but really lame.
Sorry, but not sure why im so cynical.
In fact, people don’t care about choices of which site to look at, they just want choices of homes to look at. Just tell me, why would use you over Google, if all i have to type “Bellevue, WA homes”
Mike, in many areas of the country there are multiple MLS’s that overlap area. For example, in So. Cal. there is the IMRMLS, SocalMLS, Westside MLS, Southwest MLS, CLAW just to name a few. All of these serve los angles county. Listings that appear in one MLS maynot appear in another even though they all service the same area. That is why there is 900 MLS across the country.
From a consumer point of view some broker IDX solutions are better than others. Some require you to sign in, others don’t show addresses, other more or less photos and descriptions. This is why have a simple solution to easily find the best site for your needs makes sense.
People do care about which site to look at. They are looking for the one that displays the most details and photos of homes for sale in that area. If people didn’t care there would only be realtor.com… which only has a 9% market share.
The reason you would use retrove.com vs Google is quite simple. If I do your “Bellevue Wa Homes” search on Google… I get 2.1 million results vs 193 sites that have real estate listings only.
The top ten google results have 5 real estate sites where the results are actual Bellevue WA pages. This pages hare highly SEO’d and crawler friendly which is the only reason why they show on Google results.
The rest of the results is a realtor site, an appliance sales and installation site, the convention center site, and the city of bellevue site. Our results are 10 real estate sites, with the urls pointing directly to the Bellevue search results of on each site. I think our results are more focused and targeted than those of the mighty G.
Google is universal search while we are vertical search which makes our search results much more focused. Your question is the same as asking why would you use kayak vs google to search for airline tickets?
Our results are often part of the “invisible web” which means our results are from behind the search form which most real estate sites use. This is why as you mentioned, our urls contain long query strings because that data is behind forms that typically require human interaction. Google has only recently announced that they are able to crawl behind the form.
Lastly, consumers are already searching multiple sites. They want to see FSBO’s and newspapers classifieds and craigslist and NO MLS site or other real estate site currently provides that.
Interesting post. We have domain.com.au and realestate.com.au in Australia. Domain is very accurate, especially with the inspection times and dates. They also have good photos, map options, and even street view. Realestate on the other hand can be variable. I have attended inspections which were incorrectly posted on the web.
Often, prices are not present in both websites. This is just a way of agents getting the most buyers, but its annoying to visit an inspection if the price range is way above what you are willing to pay or afford.
Hey,
There is some sort of a ” CyberWar” going on @ 1938 Media..
They aresaying that Loren is a Gangsta/Gigolo is he really working for the ” Gillmore Gang”??
A local buyer could care less that TruZillia has 3 million listings if there is only 60% coverage in their city. It is revealing that TruZ plays that gross listings perception card to sway consumers to look there– and not the comprehensiveness card.
Perhaps this post should signal a call that real estate listing sites publish their local comprehensiveness percentage (MLS coverage?) on the website so John Q. Public can decide whether he should putz around for a pad in Pomona.
These guys keep spouting transparency– it’s about time they put their local coverage stats where their mouths are.
http://tinyurl.com/3hhyau
Really interesting post and comment thread.
@ Jessie: Sorry, but I agree with Mike. That’s not a true RE Search Engine. On sites like Century21NHR.com in Seattle, we use Google’s Search Appliance to index IDX data, which makes Real Estate “googleable”. THAT’s a true real estate search engine.
We also provide our agents with access to merchandising tools that enable them to enhance their listings beyond the lame content that the MLS spits out in the IDX feed. It’s unfortunate, however, that the majority of agents won’t use them.
The one thing you never hear the RE Tech entrepreneurs talking about is how much the content we are all slinging around sucks. Compare the depth and quality of content that Amazon has around a $15 copy of Home Buying for Dummies to the average $500k home listing on any RE site, including ours.
We techies like to think in terms of data — comprehensiveness, amount of peripheral data like school info or market trends or whatever, and adapting social media to real estate. We fixate on these things b/c we understand them and know how to manipulate them.
But I submit that the average consumer would trade all the bells and whistles from all of the RE sites mentioned here for a Craig’s List clone that mandated deep, rich content at the individual property level. You know, the thing they ACTUALLY care about?
Not to mention that this is exactly what they are used to when they are researching any purchase except the biggest one of their lives.
Until brokers understand the advantage they would gain by producing that content, we are all going to be hamstrung by the fact that all we can really do is gild the lily of user experience around the rotten core of lame content that is produced by agents whose qualifications to hold their job include ownership of a #2 pencil and an opposable thumb (shout out to Kris Berg).
@John. No need to be sorry, conversations around opinions makes the world go around and helps spread knowledge and ideas.
I believe the definition of a “search engine” is program that searches for information on the “World Wide Web” not just the information on your own website. Yes, you have a very nice “search function” that uses Google search techology to sort the IDX content on your own site but I don’t think you are true “search engine” either.
I 100% agree with you regarding the importance of content and your analogy to “Amazon” but you are right, the agents will not take the time because the average listing agent is focused on marketing the property to other agents via the MLS not necessarily to consumers directly.
So if that is a problem as you say… “lame content that the MLS spits out in the IDX feed” then it would seem logical for a user to search different RE websites (which they do) as each website may present a different snippet of information or photos about the same property, which helps the consumer get more information about an individual property. Also, on your site you are still missing 20% or more of the inventory which is FSBO, pocket listings, etc… so consumers are still going to use another site other than yours.
If comprehensiveness was not an issue with consumers, then REALTOR.com would have a much larger market share than they do currently. Even users of HAR.com the public facing Houston MLS still visit multiple sites as per the WAV group report… and we make that process easier and faster than any other real estate site.