Realestate.com.au released their statistics on the prior month as they always do but something was different this time. It wasn’t that it was a record month, they have claimed many records in past fact sheets.. highest visitors, highest gap, highest percentage… You name it – they have found every way to claim they are number 1.
What is interesting is in this latest email they have ONLY compared themselves to Domain.com.au. In the past they have always included at least one and sometimes up to to 3 and 4 other real estate portals. Here is the last 5 months results but everyone I still have access to is the same.
You would have thought that Google’s entry to the market is the hot topic so they would have tried to claim a dominance over those that uploaded to Google rather than pick on their old foe.
Why is this so significant? Because realestate.com.au and domain.com.au are the only significant sites not uploading to Google Real Estate and that has been the “new phase” over the past month.
Could it be that some of those portals that jumped on the Google opportunity have started to show results already and they don’t want to flash it around?
Unfortunately I don’t have access to any July stats so lets just a take a look at June on Google Trends.
As you can see realestate.com.au has remained fairly flatline. This is typically what you would see over any 30 day period as trends in website traffic are normally more evident over longer periods but lets take a look at myhome.com.au
There is a clear trend upwards here so they entered July on a significant increase in the prior 4 week period. The beginning of the month the average was about 2.4k unique visitors per day and by the end of the month the average was about 3.3k unique visitors per day. This would represent an increase of 37.5% in just a few weeks. I am not aware of any special campaign or promotion that myhome.com.au has been running so this must be a natural increase as myhome gains further popularity.
Now Realestate.com.au is going to argue that their traffic is massively larger and increase or not it does not matter to them. It certainly is but you have to view everything in perspective. Their traffic is certainly about 40 to 45 times greater but they have to share that amongst far more properties and far more agents. Now realestate.com.au claims that Google has only 20% of the total listings available in Australia. When you consider that Myhome is only one uploader to Google they obviously have far less than 20% of the property numbers of realestate.com.au.
So if they have 1/40th of the traffic being distributed amongst just 1/5 of the properties it means that your traffic per property is at least around 12.5% of realestate.com.au. That gap is actually not as far away as REA would have you believe and if Myhome can continue that sort of growth then it will close significantly. If all of the free options start making up ground its going to make an interesting marketplace over the next 12 months.
It will be interesting to see what adding the Google factor into July’s results did for Myhome but since REA did not want to share, we will have to wait a month.
On an associated matter, I really cant stand the use of the word property seekers. It implies individual people which in turn implies that nearly 5 million Australians looked at realestate.com.au. In fact nothing could be further from the truth. Their monthly statistics are formulated by adding the daily unique visitors. If somebody checks realestate.com.au every single day from their home computer whilst they are looking to purchase they could represent as much as 31 of these “Property seekers”. If they check it from home and from work then they could represent as much as 62 of these property seekers. Since it is daily uniques that is what they should quote. If they want to quote property seekers for a month then they should quote unique monthly visitors. Because some people will visit daily, some will visit a few times a month and others will visit from work, home and mobile… my gut feeling is that will be much closer to 400-500,000 people.
************************************************************
UPDATED
I have had some interesting feedback from realestate.com.au’s PR department today questioning my take on how the 4.8 mill in the last paragraph is calculated. The state that the 4.8million is monthly unique browsers but the catch is I have to take their word for it for now but they are trying to find something to confirm to send me so we can clear it up. If they were to quote the figures the way I have suggested it would be more like around 12 million. I suggested she post her response to the article.. but she declined.
For those that use Google Analytics it now shows something called an “Absolute Unique Visitor” figure for the report. If you click on that report it will show you the unique visitors for each day that make up the report but if you add up the individual numbers for each day they do not equal the total for the month of the report.
Google provides a definition though:- Unique Visitors represents the number of unduplicated (counted only once) visitors to your website over the course of a specified time period. A Unique Visitor is determined using cookies. The Absolute Unique Visitor report counts visitors to your website (counting each visitor only once in the selected date range).
Why do I think that they haved added up the unique browser for each day? Two reasons :-
The number they quote is just unrealistic. If you start with the Australian population and remove all the kids and all the elderly then 4.8 million would have to represent close to 40 or 50%. Ask the next 10 people you speak with outside of real estate and see what percentage has used realestate.com.au in the past month. This article covers the same argument but looking at the leading newspaper sites. http://www.crikey.com.au/2008/05/26/are-nielsen-net-ratings-bunkum/
This has been discussed several times before and I have asked for more stats from realestate.com.au before on this blog and we are never provided with anything more than these huge numbers churned out by the PR Dept.
HOWEVER!!!
How the stat is calculated is actually not the main point I was trying to raise and is essentially irrelevant. Realestate.com.au uses the word property seekers which gives a misleading impression that 4.8 million real people visited the site. No matter what the answer turns out to be it will never represent 4.8 million people.
Why do the do that?.. The excuse offered was because the stats were “dumbed down” for agents. Despite the terminology used I dont have a problem with dumbing down of statistics as long as the essence remains the same same.
Just so its crystal clear Realestate.com.au is number one by a clear margin. I do not dispute that and I dont think anybody would. They simply do a far better job than anybody else right now, but that does not mean I have to accept the job they do nor any statistics that they quote.
The relative relationship between the traffic of the top portals is probably represented pretty accuratly by the Neilsens figures. No dispute here at all. But whether you dumb down the data or not suggesting 4.8 million people viewed the site over the period of a month is misleading. They should quote their stats and not make them out as something they are not.
The method the data is collected has severe accuracy limitations. Those limitations would apply across all portals of so a common error factor would mean that the relationship still remains the same. ie. realestate.com.au is kicking butt in most markets.
Why are the figures not accurate? In 2005 Jupiter Research found that 34 percent of “newbies” said that they’ve deleted cookies themselves, with that number reaching as high as 60 percent for experienced users. How does Neilsens track people.. through cookies!
How this relates specifically to Neilsen NetRatings statistics is discussed in more depth here
http://www.siliconvalleywatcher.com/mt/archives/2005/04/the_emperor_has_1.php
In short.. the Jupiter report found that “17 percent of consumers delete cookies weekly, 12 percent monthly, and 10 percent daily — behavior that “cripples sites’ ability to track users and make critical marketing measurements.”
Whats interesting is that Neilsen’s own report on the issue found a cookie deletion rate that ranged between 7% to 50% with the average visitor to Google deleting at a rate of 25%. It’s even worse than that as the cookies that are targetted the most for deletion by antispyware and antivirus solution are specifically tracking cookies… exactly the type Neilsens use.
So the company that realestate.com.au dumbs down stats from to create a “property seekers” number admits that it is up to 50% of its visitors could be counted more than once, and an external company believes the rate could be as higher as 60% for some user groups and that in a worse case scenario a person could be counted every single day (10% that delete daily) and antispam and antivirus software is even in more wirespread use today than it was back then.
The stats can be used to compare against other companies in the same report.. but you cannot convert them into “Property Seekers” because it sounds good from a marketing perspective. It just does not work that way!
20 Comments
Graheme
I understand your theory behind a person being measured as a unique browser more than once in a month if they access different computers or IP addresses. However I am surprised that you have stated monthly UBs are measured by simply adding daily UBs together, this is completely incorrect.
A monthly UB, is simply that, a monthly UB. Once an IP address has access any site within a month it is not counted again within that month. I am sure the people at Nielsen would gasp in horror at the idea of monthly UBs being calculated by adding daily UBs together…
snoop
I would suspect myhomes increase is from google re as a small increase in Homehound .
Still very low traffic in the scheme of things insignificant
Glenn Batten
Snoop.. you might be right.. it could be from Google.. but it would not be from Google’s real estate section as these are June stats not July..
As you suggested the numbers are still low.. which is why July’s stats will be interesting to see if they have kept up the momentum, especially since they will include any traffic from Google.
I can’t see there will be no miracle shift in market share but I would think there will be an impact.
Shane Dale
Snoop, I could say you are looking a gift horse in the mouth! – the site is FREE – our role in the industry is to provide an alternative and keep pressure on the main portals as we grow. Google is a perfect example of us versus REA and domain – we load they don’t – the benefit to members is there and its still FREE and will continue to be.
Currently we don’t have any marketing at all to drive this extra traffic, its all from google. That may change in the future as we change. I agree the traffic isn’t enough to seriously worry REA or domain at this stage, ….. yet. However without Myhome or homehound – you can imagine the glee and confidence with which they would raise fees.
Frankly traffic is a weak metric, enquiry level per dollar spent is the only real metric you should look at. In this case Myhome provides exceptional results for no risk to the agent.
It cannot be that hard to measure calls (using a specific number) and emails (using the source address) to show true enquiry levels – who cares what the traffic is if it doesn’t actually equate to enquiries! I prefer a traffic of 1000 hits per day with 100 enquiries to a million with 10. Quality is key. Expense/value is the other factor, you are all running businesses after all.
Just my 2 cents!
Rusty
Just doing my Friday afternoon search for properties in the Sydney Inner West suburb where I live. Thought I’d do a bit of a test to see how the different propert search portals compare. Checking across the portals there appear to be 17 properties for sale currently that meet my criteria. Domain has 15 of these listed, Realestate.com also 15 (although 1 is different), Google have 18!
Wait a minute. 18? Oh that’s right. There are only 17 for sale remember. Closer inspection shows me that Google actually has 11 of the 17 (with 2 duplicate listings and 5 out of date listings – tks Shane). Seems there’s a bit more to this listings game than those cool kids at Google anticipated. Back to Domain for me
Stuart Pike
I just wanted to wade into this conversation and clarify two things from a Nielsen perspective. I work for the company and I am speaking officially for the record.
Firstly, there is nothing secretive about any of Nielsen’s data despite what is alluded to in this blog post. We regularly deal with journalists and bloggers to provide them with figures for their articles and all anyone in such a position usually needs to do is contact us. If anyone wants to validate a client’s stated figures with us we are more than happy to oblige. You just need to ask.
Secondly, the official metric that Nielsen uses to determine rankings in Market Intelligence (our cookie centric product) is actually “Average Daily Unique Browsers”. The reason we do this is because over the period of a day the factors that distort cookie figures (and thus unique browser figures) such as cookie deletion and multiple users per PC are minimal. This means that for such a short timeframe it represents an acceptable way of measuring reach.
The reason we go for the average of all daily figures within a month is that we feel it is a fair representation of how well a site has maintained its audience over the course of the month which is usually what most people want to know.
Glenn Batten
Stuart,
Thanks for your comments.
Just so it is clear at no stage was I after any figures. I have simply asked for a definition of the figures that REA uses in its marketing. Surely they should have been in a position to advise of what exactly that figure represents since they use the numbers in their own marketing.
Also, I don’t think your rankings are in question at all. Your ability to rank the performance between clients that pay for your service is unquestioned.
In fact as I have stated from the beginning the issue is not about the rankings but how Realestate.com.au (and others) extrapolates your Unique Browsers to become Property Seekers.
Neilsens does not provide information to any clients quoting “Property Seekers does it ?
Back to how the stats are calculated. I thought it was Daily Uniques and Realestate.com.au has claimed that it really is Monthly Uniques. I am still a little confused as now Nielsen on the record are quoting Daily Uniques … but just chucking an “Average” in front of it.
It sounds to me from that as though we were both wrong but I was a lot closer.. ?1?!
Given the fact that your market intelligence uses Average Daily Unique Browsers would you like to comment officially for the record what the relationship between this number and real people or “Property Seekers” would be.
Cookie deletion is a real and valid issue relating to accuracy of all analytics. For ever Neilsens rep claiming it is a minimal factor I am sure you would agree there is somebody else claiming it is a major factor.
Glenn Batten
Hmmm..
Seems nobody wants to comment “for the record” on the relationship between “Property Seekers” and “Unique Browsers” or even “Average Unique Browsers?
snoop
I would like to see a breakdown of portal traffic by buying and renting.
My suspicion is it must be at least even.
Actual property transactions this year?
4.5m unique browsers to rea and about 1.2 for domain?
so say 6m ubs per annum into ? actual transactions?
what are all these ubs doing?
snoop
The other question is agent uploads making traffic counts inflate?
9000 offices updating and checking listings everyday ?
Robert Simeon
Snoop,
I was told sometime ago that REA count daily visitors so – yes they count the agents uploads which is much greater than if they counted weekly or monthly as repeat visitors then would only show up once. Then again if you look at the BIG picture and observe why REA and Domain have refused Google access to their portals tells a completely different story – running scared! Egos are making the decisions not grey matter.
Robert Simeon
The Google Monster – Great read 🙂
http://business.smh.com.au/business/google-monster-20090814-el49.html?page=-1
Glenn Batten
Thanks for that Robert… That is a really great article and about as balanced as you could expect to come out of fairfax at the moment.
snoop
Real estate failed in the US for google and Aus is a last ditch attempt for them ,as we are such a property centric country.
It will fail here or just be more noise at best.
Simon
Glenn,
We’ve never taken much notice of these emails from REA because , as i believe Robert mentioned, we were always told that REA record UB’s daily, whereas Domain do this monthly.
EG: I look at one listing, once a day, for 30 days in August.
On REA, i am counted as 30 UB’s.. !!
Same scenario on Domain, and i am 1 UB.
Can you confirm this via emails with PR from each company?
I prefer Domain’s way personally, and i believe that if REA were regulated and made to report ACTUAL UB data, their numbers would halve, easily.
Thanks for the great read.
MC
Snoop, America is a shocking, rambling, mess of a property market! Just look how their listings are so disorganised around so many websites – to say nothing of the shocking agency websites they have! Australia has always been so far advanced by comparison, especially on the web.
Glenn Batten
Snoop,
What makes you think that Real Estate has flopped for Google in the US??
Hitwise reported last month that Google Maps sent 2% of its US traffic to websites in the Real Estate industry, making it the #19 downstream industry (among more than 160).
Because listing data is so easily accessible in the US there is no one significant force. Even the market leader realtor.com only accounts for around 7% marketshare. You then have the other side of the coin where Google Base data of the listings is mashed up into a range of other portal sites. Certainly Yahoo! Real Estate, Realtor.com and Trulia all accounted for a larger volume of traffic to real estate listings than did Google Maps but I would hardly call it a failed attempt.
Because there are so many portals and real estate sites individual agents find they cant compete and its easier for them to just use the portals. This as MC points out has created a bit of mess with very few creating quality sites.
Regardless of all this the Australian market is so different. Google has a 60% market share in search the US but a 90% here in Australia. Real Estate traffic online is fractured into so many market players. In Australia the top two real estate portals have a massive market reach. As a comparison of relative market strength REA probably has the same marketshare percentage in Australia as the top 20 or 30 players in the US combined.
The problem for the top two portals is that Google’s sharing of data can fracture the Australian market. Maybe not to the extent of the US … but certainly more than REA and Domain would like.
Robert Simeon
Maybe Snoop did not read the article (link above) about the Google Monster – there is not a single business in Australia that would not love the traffic that Google delivers “With nearly 10 million visitors a month Google is the most visited website in Australia. Nine out of every 10 searches made on the internet is through Google. Those eyeballs translate into an estimated 90 per cent share of search advertising – the fastest growing area in online advertising as the number of advertisers using the service soars close to 50,000 in Australia. Google’s revenue is estimated to be $700 million and fast heading towards $1 billion as more advertisers divert their budgets into a medium that delivers them measurability and sales leads. Soon Google will have the ability to sell and serve richer display brand ads on 62 per cent of Australian websites.”
Little wonder REA and Domain have been left scratching their respective heads trying to work out a competitive strategy. I very much look forward to the REA response that as the article suggested is not that far off. In the meantime all our properties are up on Google Maps linking back to our website so no complaints here 🙂
Glenn Batten
Simon,
We have realestate.com.au and Domain both using the term Property Seekers with their marketing although I think realestate.com.au tie it more to the Neilsens numbers. We dont get a lot of Domain emails since we cancelled them after they wanted to increase our subscription costs by around 140%.
I have an email from Zoyee Kartalis, Public Relations Manager for realestate.com.au which confirms they use Neilsens data and states “We report on monthly unique browsers, not daily unique visitors.”
but in what appears to be a contradictory comment we have Stuart Pike from Neilsens saying “the official metric that Nielsen uses to determine rankings in Market Intelligence (our cookie centric product) is actually
Stuart Pike
Sorry for the delayed response.
I can’t really comment on the relationship between unique browsers and property searchers because we don’t measure or report the latter.
We use average daily unique browsers within Market Intelligence as the metric if choice for ranking etc. as this use of a daily time period mitigates/minimises cookie deletion. This means the daily unique browser figure represents a reasonable approximation of a people metric (i.e. probably out by 2-3%).
Cookie deletion is a significant factor in any measurement based on cookie data. It is the major reason why the use of monthly unique browser figures is not encouraged by Nielsen. That is not to say we do not report the monthly unique browsers for clients (for some it is an accurate figure due to their low frequency of visitation which means cookie deletion doesn’t have an effedct) it is just that we acknowledge they are potentially overstated (especially in websites that have frequent visitation) due to cookie deletion and use from multiple locations. I would hope that there isn’t a Nielsen employee on the planet that is saying anything different to this.
Regards,
Stuart