The new web design from Realestate.com.au was built from the ground up. Of course when you do such a major rewrite of the code you have the opportunity to massively improve certain aspects of the site by wiping the slate clean and starting again.
The web analytics for the portal has been beefed up and we are about to have a lot more data to look at. With the old versions we effectively had page views and enquiries.
One of the most noticeable impacts the new site had was on the number of property views to each property. Overnight property views dropped by around 30 to 40% across the board. This was in part due to the new user interface which allowed visitors to see more of each property without browsing on the details page. Without access to page times and other data there is no way for an outsider to know exactly how much this played in the number of page drops but account managers have been using that excuse when copping flak about the drop in activity from clients.
Mark Twain once said “Facts are stubborn things, but statistics are more pliable”
So the new weapon against agents expectations and justification of the ever increasing cost is the term “Engagement”. Better get use to it!
So accordingly to REA property views are no longer relevant and will no longer be used because “measuring by property views is too limited in light of the new richer ways consumers can interact with your listings and your brand”. The measurement and benchmarking of agents in the future will be based on “total engagement” numbers.
With the focus on engagements rather than property views it means we are not going to be able to quite so easily compare against statistics to last year or even to other portals.
So what do we know now?
There are 21 new sets of statistics across four categories as follows:
Views
- Property Details Pages
- Property Carousel Views
- Property Map Fly-outs
Engagement
- Property Photo Views
- Map Interactions
- Floor Plan Views
- Virtual Tour Views
- Property Video Views
- Street-View Views
Saving and Sharing
- Saved to MyRealestate
- Notes in MyRealestate
- Print Page
- Send to a Friend
- Share on Social Media
Enquiry
- Emails to Agent
- Agent Details Views
- Click-throughs to Agent Website
- Inspection Times Reveals
- Inspection Times Saved
- Auction Times Saved
- Property Directions
I have also been provided with a snapshot of real data that provides some interesting early perspective. This sample is well over 100,000 property views so it should give a great indication of what to expect when it gets rolled out to everyone.
- For every 1 property view there is 7.4 property photo views in the engagement category.
- Property Carousel views are 11.5% of property page views but only 1.5% of property photo views
- Map engagements run around 2.6% against property detail page views and from this data it shows floorplans were far more popular at 3.6% Remember, most properties have a map, but very few have a floorplan so to me this highlights the importance of a floorplan.
- Street view engagement were the least popular at 1.6% of property page views
- Saving and Sharing was very very low and all 5 types combined only added up to 2.5% of property page views.
- Enquiry statistics were even lower totaling around 1.04% of property page views. This was made up with a little under half of those numbers by email (0.45%) followed by inspection times (0.20%) and property directions (0.15%).
I think the level of detail is admirable and it will be interesting to watch how those number relate to our own agency. Right now the markets in different parts of Australia are completley different so there should be some major differences between agencies in each state so the numbers quoted above may have little or no relationship to what you might see for your own account.
23 Comments
Nick
I think when you are paying for a subscription, you are entitled to know exactly what you are getting. Statistics like these are important, but they can be smoke and mirrors.
The focus on engagements is interesting because while they are interesting to know, they dont actually mean much in the context of selling a house. E.g. the number of times street view is seen does not help you sell more houses nor does it justify the subscription at all.
For Luxury Homes Australia, awhile ago we decided that we must give detailed statistics to our clients to show them what exactly they are getting with us.
Here is a scrubbed screenshot of someone’s stats:
http://www.luxuryhomesaustralia.com.au/lha_stats.png
Note: User specific data like exact numbers has been obscured. Subscribers get exact numbers. There are also graphs for individual properties.
All the statistics are live. The bar for the 25th is small because its early in the morning.
We decided it was useful for our users to know where the visits were coming from, and how often their properties were seen (e.g. in search results) so we give them those details.
Sal Espro
Thanx for the update Glenn. So, ‘going forward’ we are to investigate ‘Engagements’, eh?
How about I start a website comparing enquiries gained from each of the advertising portals and agents feed in the stats? Better than Neilsen ratings and far less duck-diving, bullshit, excuse-making spin from REA, Domain, REView, etc etc ?! How about we wrest-back the control.
Glenn Batten
Nick
I think we all expected there was going to be more detailed stats package provided but honestly I was blown away by all the options.
That that have a better understanding of analytics will be able to pull some great analysis out of it. To most people it will just be a bunch of cold numbers!!!
What fits into what category is interesting.. How Inspection Times Reveals is an actual Enquiry for instance..!
So all in all I think the stats themselves are a great thing. We should be able to make some changes to our content styles and get a good indication of what effect it has.
The next thing will be how the marketing team plays, spins and spews out the numbers that the tech team is collecting for them 🙂 They may get interesting!
Glenn Batten
Sal,
There are quite a few people who hold that information for their agencies tracked over many years and it provides an interesting read… especially when you graph the cost of the subscriptions against it.
What I found really interesting was Peter’s article from June 2006 that quoted “If you look at the pricing of today for a basic subscription to realestate.com.au which is $379.00 per month, at the current rate they increase subscriptions each year it will be $515.61 per month by 2010.”
Well we all know that it has far exceed that!!
Jayne Harwood
I am also struggling to get my head around engagement!
We market ourselves as the Number 1 agency on Real Estate.com for most Property Views in Tasmania which we have been for the last twelve months. And potential vendors love to know we are Number 1 on the Number 1 Real Estate Portal. Since the launch of the new site our Property Views have taken a hit and I am advised to now focus on engagement!
In May for the first time ever our own website http://www.bushby.com.au received 68,000 property views which was 10,000 more that Real Estate.com. Apparently we were the Number 1 agency for “Engagement” in May so why do I find myself wishing for the old days where Property Views, Email Enquiries and Click through to Agent Details was all I needed! Now I have over 20 different things to analyse. Of course I can still get the 3 KPIs I like they just gone down in number….
PaulD
I agree Glenn, the statistics are a good thing. It’s a pity many agents will probably not use them. However, that is the way it is. Statistics can sometimes be a bit like knowing the displacement and weight of the pistons in your car. Nice to know – but all you really care about, is that the car gets you there.
In relation to REA price increases – in the case of my office the increase since 2006 has been an average of 19.5% per year over that period, and in fact it has exceeded 19% increase p.a. since 2003. In a business that is absolutely constrained by the vagaries of the property market in this country, REA will have to keep trying to invent new ways of not providing any additional information, and charging more and more. That task will get more and more difficult as competitors are able to muster support from the agents due to an ability to offer a similar service at a lower cost. People will say – yes, but they won’t be able to find every listing in Australia on the same website – To that I would say — So What ?– people are not that incompetent that they can’t find listings in an certain area or on another website, and obviously that are already doing that.
Peter Ricci
1. Just because a site tells you to look for some certain stat – does not mean that stat is relevant. Does having the number one spot for a suburb mean that listing will sell quicker? No, and there are no stats that back this up, does this mean it will get more views – of course it will.
2. To your vendors the only stats are relevant are conversions to sale or rent, unfortunately some also want the feel good factor, but people like that are easily fed and rarely pleased.
3. As for prices the portals charge you – to me pricing is not the BIG deal, it is all about what they CAN charge you and the power they have over you.
The best example of this is Telstra and line rentals. Homeowners have been paying $$ per month for decades to rent a line from the street to your home. that would cost less than a dinner to supply. So how can they do this? Because you have NO choice, they own you!
This has always been my point, you take – or distribute the power away, you will find prices will steady (even decline), innovation will increase and they will approach you on bended knee.
This is why competition is good for everyone!
Glenn Batten
PaulD,
If we can get this detail down to a property level it should open up a good opportunity to do some split testing
In the past REA staff have contradicted themselves over which provides better results. Bullet points or free flowing text or combinations of both. This would allow agents an opportunity to do their own testing for your own local market and feel pretty confident about. Testing different call to actions.. whether internal shots or external shots as the main photo convert better. Specifically mentioning the open homes in the body text. There is quite of lot of things that you could test..
Glenn Batten
Peter,
1. I think we are all pretty immune to the pointless stats issued by the portals now. This is why this new level of detail is impressive for those that can use it. To everybody else it will just be a page of meaningless numbers similar to how the $6 million unique browsers has no relevance to an individual agent. In fact notice how they dont provide agents with unique browser counts at any stage?
2. Agreed, but thats an entirely different thing. We are talking about stats a portal gives an agent, nor stats that an agent would give a seller. If you went to a seller and spoke about engagement rates with a vendor you could expect to be shown the door. Horses for courses! To many sellers the most important stat is even simpler… how many signs they drove past on their way home that your sold sticker on it.
3. Certainly the power they have is the real problem and the prices are just a symptom of the problem but to the thousands and thousands of agents actually paying their bill every month its the price thats the most visible. When you have a cold you dont bitch about the little bugs running around in your bloodstream, you bitch about the sore throat, the headache and your runny nose. The fact that they are just symptoms of the real problem does not make you feel any better 🙂 The patient worries about the symptom and the doctors worry about the underlying cause. The question is who are the doctors ??
Jack
Thanks for sharing up
Glenn Rogers
The only relevant stats to the agent are, in my opinion –
Listing views – when your listing appears in the search results.
Property views – when listing clicked on, in full view.
Click through to agents web site.
Subtract from this search engine spiders and stats from linked web sites , such as your own web site if it’s linked through to the portal and thats it.
The other stats such as photo views, carousel views and the rest should be analyzed by the portal to determine what’s attractive and what isn’t , whats easily seen and what isn’t, what needs to be given more prominence and to make adjustments to the site and tweak where required.
Mac
Totally agree Glenn. Enquiries are only going to result from properties visible and clicked on by buyers. The rest of the stats may well assist in increasing enquiry rate but the absolute number is going to be directly related to property visibility in the first place. Glenn B, Robert et al, what sort of stats do you get for ‘Enquiries per cost of portal/advertising destination’? (I know Domain is dominant for you, Robert, but would you mind saying to what extent?) e.g. 200 equiries per $450 pm fee = $2.25 per enquiry. (It’s a bit like CPM for banner ads, isn’t it 🙂
Glenn Batten
Mac,
The variables that go into any cost per enquiry calculation mean the figure alone is pretty pointless. An agency running in Mermaid Waters on the Gold Coast might only have a dozen properties all over $5 million each. Whereas another agent in Toowoomba might have 200 listings and 80% of them are all under $400,000.
I know our cost per email enquiry (not ready to believe that an inspection time reveal is an enquiry just yet Mr REA!) has skyrocketed as the market here is is far far from what Sydney and Melbourne are seeing.. We have gone the other way and cost per email enquiry would be double that of a few months ago…
I have had a play with the new stats dashboard (check your REA backend).
You have to burrow down to get the full details. On the surface they just show the category names (Views, Engagments, Sharing and Enquriy) and most agents will assume that enquiries actually equals email enquries and that not the case. Their old numbers are just a subset of the new numbers…
The dashboard is very buggy and it is crashing alot.
Nick
Oh are they hiding the inspection clicks under enquiry? Thats downright misleading.
Agents that dont look too closely will take it at face value, and some who dig a little but dont pay attention to whats happening will wonder where all their emails are going.
I just had a interesting idea however. I’ve recently done a extensive integration of a client’s website with Google Analytics. Any tech people here should check out Custom Variables. You can create some incredibly powerful stats with them, for example the client can now get a full breakdown of search behaviour on their site.
Now if you could present that data to agents on a portal….
Mac
Quite agree, Glenn. However, as an industry index I think that the cost per ‘actual’ ‘contact’ enquiry would be very useful. I don’t think this is affected by whether you have 5 x $10 million homes or 8 x $250,000 homes. The enquiry is why an advertising destination is primarily used and so the cost/enquiry is very relevant. Perhaps we could break it down to rural, regional and metro, but it is an important industry-wide set of stats to know.
Glenn Batten
As a macro stat looking at whole districts or states to set a baseline I would agree. The danger in looking at individual offices is that another office can look at it and say my figures are better or worse and really not understand why.
As an example using real stats why individual agent results are unreliable is it also depends on what product that agents utilises.
Looking at one agent in the statistics I quoted in the article…. They use a lot of add on products and they certainly generated a bucket load of email enquiries. Given what they spend if you used email enquiry only as a base metric it was costing them the equivalent of $45 per email enquiry over the past month.
A simplified view I know because they also generate phone calls etc etc.. but it is the sort of numbers we are talking about using for comparison.
If they just had a standard platinum subscription they would obviously expect their email enquiry to plummet…..
so statistically if their email enquiry dropped by as much as 50% in that case this means that the first 50% of all enquiry per month were priced at a little over $8 and the second 50% was costing them over $80.
If the drop was more like 75% then the first 25% of the enquiry cost them $16 and the last 75% cost them $55 each.
The problem is without all that extra information how does another agent use it as a comparison to theirs?
Of course, to the agent themselves this data is critical. If I was them I would be asking myself if the extra $39 to $72 per enquiry would be able to generate more enquiry and better spent elsewhere but that is a whole other discussion and the decision itself certainly involves more than these numbers anyway.
Mac
Aren’t they VERY interesting figures, Glenn! This is exactly what we need as an industry. (And why couldn’t they be provided anonymously).
Sure, we then need to look at how qualified leads from one source compared to another might be e.g. it might take 20 enquiries to sell a listing and how many of those are well qualified will affect the cost per qualified enquiry. However, in the first instance, to know that it is costing $45 per lead from one source as distinct from $20 from another and $175 for another would provide useful metrics if you had to choose between one advertising destination and another e.g. Do I add REView to my ad list while keeping REA and deleting Domain etc etc? (And I would think that any new portal model e.g. MyHome might be able to swing some business their way once these stats became more transparent).
Glenn Batten
The free portals seem very cheap in comparison 🙂
Every agent needs to do these sort of calculations for their own office and if they do i reckon they might look at their eMarketing a little differently. Maybe agents need to wrestle back our own share of our own wallets 🙂
Vic
“Paralysis by Analysis”. This is the affliction caused by the provision of endless statistical data, from a service provider to a service user, which primarily is used to on behalf of the provider to confuse the user in to believing he/she must and is dependent and behoven to the provider.
The only real data an agent needs to know is where did the lead come from.
Vic
Glen,
Thanks for the article.
Having been at both ends of the scale ie corporate and client, I know that statistical data sets needed by both to run their business are quite different. The swathe of statistical data coming out of REA would be of little interest to me as a client (agent) in running my day to day business.
However, simple statistical data showing how many leads came from each listing would be invaluable.
The way I would like to see REA use the stats is to do analysis, and come up with recommendations to agents as to how or what they should do to achieve better leads outcomes.
Glenn Batten
Vic,
I have to agree with you…
As I mentioned earlier for most agents the data wont mean much. Some agents will be able to extract more out of it than others. You can use it for some rudimentary split testing but if we could get the same numbers at a property level it would be easier….
But best of all is if REA use this information to work out what works best and to report back to us.
What is a shame is that most agents will not delve deeper into the stats and they will take the Enquiry, Engagement and Views at face value and use these numbers to compare against other portals …… of course that would not be their goal all the time would it? 😉
Vic
Glen,
There will always be those that know how to use statistics to evaluate their listings performance and those that have no idea or just don’t want to bother. My point is that REA as a service provider, and because everone pays the same rate to them, data should be dumbed down so that outcomes are given to the client and not a bunch of numbers that need a time resource, at the agents office, to produce outcome.
For example lets take three agents at the extremes of analytical abilty:
Agent A- Takes the data supplied by REA
– identifies the leads that came from REA
– analyses the rate of activty against these leads ie number/ % of floor plans viewed, number/% of video views etc
– conversion to sales from these leads- from sales staff
This will tell this agent three things:
1.the best way to present advertisements and
2. what best practice is being used by successful sales staff
3. uses the data to follow the sales principles of a). do more of, b). do less of c) scrap altogether.
Agent B- Collects the data and gives it to his admin staff for analysis and passes it onto his sales staff
Agent C- ignores the data and expects their service provider to provide the analysis/recommendations.
I would daresay that most will fit the category of Agent C and REA should recognise this as it is this critical mass that keeps them in business.
Your analysis in the latter part of your article in part proves my point. You have analysed statistics which tell me our advice to our portal users should be:
– include a floor plan
– ramp up social media
There is much more that can be espoused on the subject of use of statistics and perhaps some “expert” could do an article of this, but for now the debate is on whether the new stats regime from REA will help the majority of paying customers and my simple answer is NO.
Cheers
Glenn Batten
I think we are both making the same points. In fact for most agents I think the stats may be a problem in that they will assume the category Enquiry is the same as the old Enquiry which is obviously not true. Some agents may be getting a false sense of security from the new stats!
With regard to the portal using the stats to find better results I remember this in another thread once before when a UK portal who found that simply changing the colour to orange I think it was on the reply form increased enquiries massively. This is the sort of thing I believe you are talking about the portal doing their own analysis for the benefit of their clients.
They now have the ability to do this sort of testing and analysis lets hope they do.