You don't know the power of the dark side

vaderAfter playing with Zillow for the past couple of days, the first words that come to mind are “Impressive, but you are not a Jedi yet”. I have a hunch the guys at HouseValues are going to get “Netscaped” if they don’t take their game to next level.

The Good
The UI is slick. The mapping isn’t quite Virtual Earth / Google maps slick but it’s close (if you add mouse wheel zoom, arrow key navigation support and resizeable maps, I’d put it in that league). Seeing all the lot boundaries displayed on the map is something that I haven’t seen done well before and is a feature that will be expensive or difficult for Zillow’s competitors to match. I like the fact that they partnered with GlobeXplorer, since I believe that will enable them to out map RedFin.

The Ugly
I find all the trash talk about uptime and availability amusing. As any experienced software engineer will tell you, the first days for any web based service that has had the anticipation & hype of Zillow are going be rough. After all, if the mighty Microsoft had troubles with X-Box Live when Halo 2 was launched a few years ago, the fact that Zillow’s first day had some minor troubles is hardly surprising. Besides, I’m sure Rich Barton and the boys will buy a few AMD Dual Core Athlons CPUs with the new WD Raptor drives during the next few days and cure that problem.

The Bad
I suspect biggest problem with Zestimates is the current lack of high quality data. (Gee, the same issue I keep complaining about). Any realtor will point out, doing an accurate completive market analysis house is a problem that involves many, many variables. I understand it’s a hard problem, but the fact that the Zetimates are so far off for my house (which I thought should be an easy case) is disappointing. I don’t expect accurate estimates for waterfront, hilltop views, high rise condos, Bill Gate’s house or rural properties. But my house is a cookie cutter house is suburbia (with lots of similar houses for sale). I would think my house would be an easy one to get right.

Just for kicks, I implemented a quick & dirty Compeitive Market Analysis feature for the Rain City Guide home search. I found it to be more accurate than Zillow for my house, and Dustin thought my estimate was right on the money for his home (after he entered the correct square footage). Anyway, play around with it and let us know how close to the “right” price it is for your area.

BTW – My version just goes against active NWMLS listings (so forget about trying it if don’t live in Washington). It’s pretty crude and it’s not as cool as Zillow, but then again, I hardly have $32 in venture capital (much less $32 million), so cut me some slack!

I think the nay sayer would be wise to recall the words my former boss once said, “We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten. Don’t let yourself be lulled into inaction.” Zillow’s Zestimates may be off in Safeco Field right now, but I have no doubt they will get much better over time. And when they do get better (not if), you better be ready.

Well, I’m going to shut up now and let my code do the talking. I’m sure the engineers at Zillow are following suit.

Caffeinated Software

12 thoughts on “You don't know the power of the dark side

  1. Pingback: Bronte Media » Zillow Launched

  2. Hey Robbie,

    I like yours a LOT better than Zillow. Yours says I gained over $600,000 in six months! I’m over a million four! Woohoo!

    Let’s suffice it to say that I could, with techie assistance, write a program that would in fact actually accurately value any home in the country. What value would such a site have? What are the practical applications that would produce income?

    Maybe there’s a market for going to someone’s home and valuing it for say $1,000 and then leaving? Real Estate Consultant?

  3. I’m glad you like the number, but is that $1.4 million figure more accurate than what Zillow said? That’s the real question. Besides, you live near Kirkland waterfront property, that’s going to screw up my simple software model (and Zillow’s not so simple model).

    BTW, there already is a market for going to someone’s home and valuing it. They are called
    real estate appraisers.

  4. No. You are both way off.

    Yours is comparing to New Homes. A brand new home with the square footage of mine and in my location would, in fact be $1.4 million. Kind of like buying a new car vs. a “used” car.

    Zillows is using tax records primarily, and houses built in 1921 with huge 1994 additions don’t conform well to the Zillow model.

    Some homes are just unique!

    Real Estate appriasers do not value property the same way as real estate agents. They tell you what is based on what was…we tell you what can be! LOL

  5. I like the Zillow guys but I think that pouring all that money into such a site was a very dotcom thing to do – webvan came to my mind.

    Most homeowners will take a look once, say “nice” and then forget about it. After all, this information is available to them in many other ways, less slickly but more accurately. If this was a home buying portal with active listings then I could see more “raison d’etre”.

    Most real estate agents will feel threatened – after all this is their turf (maybe this is a good thing, in the longer term). I can see homesellers or buyers questioning their agent as to why their price and not the Zillow price is correct.

    Most of all, to keep this site accurate, Zillow will have to spend a lot of dough, otherwise it’ll be garbage in garbage out. I’d love to see the business plan for this operation!

  6. zillow takes public information (to whatever extent available), digests it and spits out a price range. I prefer property shark’s method, which is simply to give you the information and let YOUR BRAIN digest it.
    Little known fact: Barton rooted for Deep Blue to beat Kasparov 🙂

Leave a Reply