Friday Flash: Frontdoor, fuzzy housing math and live burrito-cam

Frontdoor.com rocketed to #2 on the Hitwise version of online real estate’s traffic numbers for January. This was the result of a sweepstakes run by its corporate cousin, HGTV. A few things to note related to this:

  • Frontdoor is an excellent site that has trailed online real estate leaders in audience for the past three years, but nonetheless remains tied to a massive lifestyle media machine. If the powers that be at Scripps Networks ever decide they want a category-leading real estate site, they could make it happen almost instantly.
  • We in the industry remain focused on traffic numbers, but this spike highlights an important question: what is a visitor really worth? Is a sweepstakes driven visit to Frontdoor.com worth as much as a direct visit to Realtor.com or Trulia? Is a visitor who ends up on Zillow from a Google search for “Utah Real Estate” worth a visit to a local broker site driven by long-term brand association? Of course not. If you’re a broker, you need to look beyond the big numbers and top 10 lists to ask where, exactly, they came from.
  • Scripps Networks Interactive’s stock hit a 52-week high this week.

Factual, one of several new geo/local data companies, announced an enhancement to its API and a new iPhone SDK, making it easier to integrate point of interest data into web and mobile apps. Factual is free for the time being. The other interesting thing about Factual is that its API is read/write – meaning end users of apps that integrate Factual can edit and enhance the data set.

I’ve never thought the capacity to expose the context around a property elegantly has been fully met. I hope developments like this allow us to get there.

My first thought when I read that Nordstom bought an online flash sale site for $270 million in stock was “Why didn’t Realogy buy Trulia 3 years ago?

Matt Carter at Inman wrote a great piece explaining a report from Core Logic claiming that NAR’s tally of existing home sales is 15-20% high. It’s shocking that this is not shocking. But what really jumped out at me was one of the entries in the Q&A doc NAR issued in response:

Q: How are NAR home sales computed?

A: NAR collects sales data from numerous MLSs, with a reporting sample of about 40 percent. If data computes to be a 5 percent increase from one year ago then we say home sales rose 5 percent from one year ago.

I learned all about predictive models and regression analysis in school, but how is it methodologically sound to sample 40% of a highly localized and variable market? Can someone explain this to me?

Now if we only had a national property database …

Google made search more “social” by integrating twitter, Flickr and Quora into results. This is handy. But something like this will never materialize in real estate search. Engagement with real estate online is generally episodic – meaning, for example, that while I will always see Facebook friends woven into my experience on Yelp because everyone always eats, I am unlikely to get much help from my social graph when searching Zillow. Just not enough people engaged at any one time.

But imagine if we hand-picked members of our social graph to join us on our online real estate journey across several sites? Facebook almost allows that right now. It’s going to happen. Start thinking about social real estate search now.

If you’re like me, you sometimes have pangs of self-doubt. You wonder if you’ve made the best choices in life. If you could’ve done better somehow.

Well here’s a little pick-me-up…

It seems that at the North African-themed Sahara Hotel and Casino in Las Vagas, there is a NASCAR-themed restaurant in which one can order a 6-pound burrito. If you eat all of this burrito, you don’t have to pay for it. But the truly wonderful thing is that your effort is streamed live on the web (scroll to the bottom of the page).

I watched three guys go at it for about ten minutes this morning then went back to work, self-assurance restored.

Happy weekend!