Scraping Property Listings

A few months ago a client of mine received an email from a web application called myhousehunt. The site was scraping property listings from the major real estate portals and populating a portion of each listing on their website. They were then allowing property seekers to build a timetable for open inspections which could be printed or emailed to the user’s phone.

When i first looked at myhousehunt I noticed it was displaying listings from REA and Domain. I thought this was unusual as I doubt they had permission from either portal to do this. I recently visited the site and noticed they explicitly state that:

“MyHouseHunt will search for properties currently listed on Domain.com.au”

So it appears Domain have given the green light while REA in typical style, have unleashed their lawyers on the scraping site.

I’m always puzzled why REA will never share their content. What threat does a site like myhousehunt have on a real estate portal, in particular REA? Nothing, they display a few details of each listing with a link back to the original property on REA. All this will do is strengthen the REA brand and drive traffic back to their portal increasing those UB’s they’re always talking about. Thumbs up to Domain for embracing applications like this as it encourages greater innovation and the creation of cool and useful real estate applications.

MyHouseHunt, Real Estate Applications, scraping sites

SEO For Real Estate
Listing Leads
Agentpoint Real estate

About The Insider

Ryan has been involved in the real estate industry for a number of years. During this time he has operated web based real estate businesses along with provided consulting to the real estate software development industry. Ryan operates Agentpoint.com.au and is actively involved with the design and development of real estate systems, software and web sites in the United States, Australia and New Zealand.

13 Responses to Scraping Property Listings

  1. Peter Ricci December 10, 2009 at 11:30 pm #

    Here here! God to see you posting on here Ryan. I think the cartoon picture is a lot better than your real life one as most people are eating when they read this in the morning.

    On a serious note, Domain seem to be leading in this department. I still have no idea why these portals do not yet have an API, it is so closed and backwards.

  2. Alex December 10, 2009 at 11:51 pm #

    Hi Guys,

    Just thought I would let you know that Homehound does indeed have an API –
    http://www.homehound.com.au/services/

    We have migrated many a scraper over to the API and it works well for both parties.

  3. Charlie December 11, 2009 at 12:38 pm #

    It’s an interesting one; as someone who’s fought against site scrapers before (years ago, I knew not) as I felt (then) that they were not really helping anyone, but themselves and their advertisers, and pretending to have a portal, I was against them. One that was scraping me (but not REA or Domain) from 2002 through to 2005 I took as a personal attack because they had been set up as a portal in opposition previously (starting a few months after we started 10 years ago) and then remorphed into a scraper that took clients they could not win the old fashioned way (like ringing them up, winning their business, developing their site as we did)… so you can see my train of thought.

    Nowadays? I suppose Google real estate (all power to them) have changed the game; and others like suburbview scrape listings and link to the agents’ sites. So this could be the future. It’s how it works in the States isn’t it? (Zillow.com, Trulia.com, etc).

    If so, then the price of portals is going to zero (hurrah!), and agencies will just need their own website (everything points back to it) and a web developer/system (to feed you up to where you want to be thru APIs, or be scraped…). It won’t happen overnight, but … the slow inexorable shift may be coming.

  4. Ryan O'Grady December 11, 2009 at 12:55 pm #

    Alex,

    This is very exciting an Australian portal with an API which developers can access. Myhousehunt are probably still scraping, at most they might be receiving xml from Domain. I did try and read your terms and conditions but the link in the API section doesn’t appear to exist.

    So, the application credits Homehound. Can the application display all property information which is available of Homehound? Is it a requirement for the application to include a link back to the listing on Homehound? And lastly, can you provide a few examples of apps using your API.

  5. Robert Simeon December 11, 2009 at 3:13 pm #

    Great post Ryan – and welcome 🙂 REA make a habit of being outed in yet another clear example of double standards.

  6. Alex December 12, 2009 at 11:07 am #

    Hi Ryan,

    The API will return a snippet of the available data but enough for a credible search result. We would then hope the listing is linked back to Homehound so the user can see the full details of the property.

    Here is a simple example of a request for sale properties in 2015 (results are paginated) –

    http://www.homehound.com.au/api/?api_key=35bea4999ab0f86416316736493a10c4&method=homehound.listing.search&postcode=2015&sale_type=S

    Please remember to register your own API key if you decide to use the API as the example key gets changed periodically.

    Hope that helps.

    Alex

  7. Bill Burdin December 14, 2009 at 12:41 pm #

    Alex,
    Thanks for that, I’m wondering if it’s possible to call the listing agents details in the response.
    Thanks
    Bill

  8. Nick December 15, 2009 at 10:21 am #

    Alex, good work with the API, but as it stands its pretty useless.
    All you can make with it is a HomeHound clone which points everyone back to HomeHound.
    I cant think of anything else you can do with it.

    What you need to do is go all out and provide full access to *everything* for it to be any use.
    Look at say http://www.last.fm/api/intro They have everything.

    You then police its usage with a licence agreement to stop people doing things you dont want them to (e.g. must provide credit, no crawling, etc…)

  9. Alex December 15, 2009 at 12:21 pm #

    Thanks everyone for their feedback.

    The API is early stage and I am sure we will open it up once we have some policies in place.

  10. Ryan O'Grady December 16, 2009 at 6:43 am #

    We look forward to that and let’s hope some of the other portals follow your lead.

  11. Bill Burdin December 16, 2009 at 12:24 pm #

    I would suggest if you open up the API to it’s full potential clever developers would create some killer apps that would blast homehound to the forefront of real estate portals. Personally I can think of several apps that the industry would embrace.

  12. v4t October 2, 2010 at 8:38 am #

    so anyone knows that is it legal for myhousehunt to do the scraping or they have backup from realestates.
    I read somewhere that REA complained propertyguru for scraping its data before.
    A quite popular recently raising site is http://www.rbasket.com.au .Look like they scrap data from househound and hartcourt.

  13. peter November 13, 2012 at 11:28 am #

    Interesting read. Thanks to http://www.Domain.com.au and homehound API, those scrapping sites like oldlisting.com.au and http://www.estatecentral.com.au can survive. I found them useful sometimes.

Leave a Reply