WE LIVE IN AN AGE OF consumerism, where potential clients have more information about choosing an agent than ever. But it is also true that we can access a whole lot more information that enables us to be smarter at marketing to them. Together with CoreLogic Australia, Elite Agent Magazine gathered an expert panel to share insights as to how new tools are re-inventing prospecting, and why traditional coldcalling may soon be a thing of the past.
With data and tools available today, it is possible to predict with a fair degree of accuracy, a vast array of human life changes, well beyond the four โtraditional Dsโ (death, divorce, debt and departure) of real estate. These life events might include having a baby, teenagers leaving the nest, a change in jobs, trends in spending patterns and much more.
Beginning the panel conversation Georgie Brooke from Greater Data acknowledges there are several different โprobability to listโ models. โ[For example,] these guys,โ says Brooke, referring to Quantium and Data Republic, โhave an enormous amount of behavioural information, and weโve got the demographic and life stage information. You can put these together and predict when someone is likely to do something.โ
1 . BIG DATA EQUALS BETTER COMMUNICATIONThe aim of โbig dataโ services is to help you to communicate better with your current and prospective customers. Rather than calling people you donโt know โout of the blueโ, or marketing to everyone, it is possible to get close to predicting which house, in a street, in an area, is most likely to list next โ and then use smart methods of content marketing to attract them to you.
The tools we will discuss in detail in this roundtable wrap-up have been created by the teams around the table (with others under development):
A. Smartlist, a probability to list model built by CoreLogic and QuantiumB. Quantium Facebook QSegments, which allow agents to market their business or a particular property to a targeted audience on Facebook
A. SMARTLISTSmartList, released last year, by CoreLogic and Quantium looks at the relationship between property data and consumer/demographic data, with the goal of identifying properties most likely to list in an area.
The output is a list of addresses that are indicative of a vendor listing for sale, ranking those addresses according to how likely they are to sell. Clients can then use this list to market more โsmartlyโ to those addresses, with more personalised direct mail and focused door-knocking, phone calls, emails and other usual methods. Data is updated and a new list distributed monthly.
Says Annie Liao of Quantium, โ[SmartList] essentially is taking all the information we have about householders plus the 9.8 million or so properties here in Australia (through CoreLogicโs data), tying that together with all of the market trend data analysis information and consumer insights data. Then we try to predict who is more likely ready to sell through consumer behaviour and also whether the area has been good with recent salesโ. The output of this is people who are more likely to sell; a warm list of prospects.
B. TARGETED LISTINGS ON FACEBOOK (QSEGMENTS)
In the last 12 months, Facebook have made more changes to their news feed algorithm which favours content from your friends, rather than what you may be posting to attract new clients or buyers.
One of the main elements of feedback from their clients, says Liao, is not being able to ‘activate’ directly the insights that big data provides. As a result, Quantium has partnered with Facebook, as a media activation channel, and created more than 150 segments (QSegments) across consumer-type behaviour, socio-demographics, property attributes and more. Liao explains, โThis allows our customers to target specific types of people that they want to talk to on Facebook.โ
Josh Cobb from Stepps joins in. โIt’s one thing to target people based on certain age groups and location on Facebook with your ads, but it’s another thing entirely to be able to target people based on how many bedrooms their house has. Or does it have a pool? Or do they have pets? Or how long have they owned the property? Now agents have access to be able to run ads using that data set, thanks to CoreLogic and Quantium.”
2. YOU HAVE THE INFORMATION; NOW HAVE THE RIGHT CONVERSATION
So exactly how good are each of these models? Says Liao, โWith all predictive modelling there is no 100 percent certainty. Itโs about highlighting and identifying the people who are more likely [to do something] based on their behaviour.โ
Douglas Driscoll from Starr Partners has been using the SmartList product for a couple of months and is one of the first in the industry to do so. He notes that the data is impressive, but there are a couple of challenges that agents need to come to terms with as a business to get the most out of the system.
โThe [probability] numbers we are talking about is that one in 10 people will list their house in six months, and one in five within 12 months. They are pretty impressive numbers.
โBut itโs all about the approach. Youโve got a list of names and numbers, so youโve got to figure out the strategy for contacting these people. Is it a mailbox drop, is it an intelligent letter, is it an email, is it a phone call? So that is where we are still trying to come to grips with it. But so far so good.โ
Liao agrees; it is still going to come down to the conversations a real estate agent subsequently has with those prospects. โYou have to understand the behaviour behind the algorithms that identified them in the first place to increase your chances of converting the listing.โ
Just like “๏ปฟ๏ปฟLetterbox๏ปฟ๏ปฟ Dropping on Steroids”
WHILE MANY AGENTS USE FACEBOOK to reach their local community, over the last few years, notes Josh Cobb of Stepps, it has been harder to reach those people on Facebook, no matter what industry you are in, unless you pay Facebook to promote your posts.
Enter QSegments, a product created by CoreLogic and Quantium, which Cobb has been running some tests on with some of his clients. So far, he says the results have been outstanding in terms of both promotion cost reductions and website conversions and it has been referred to as “Letterbox dropping ‘on steroids'”.
โWhen it comes to content, we spend a lot of time talking about the โtop of the funnelโ: attracting people to our business through helpful content, whether that belifestyle or related articles, how-tos, or establishing someone as the leading authority in their local market.
โOn the property-related ads that weโve been running for a select group of clients [using QSegments] on Facebook, specifically promoting a property or development of some kind, weโve been able to reduce that cost for the same result by up to sixtyper cent,โ says Cobb.
โWeโve seen up to a 70per cent decrease in the cost perconversion, because what weโre doing is essentially being more relevant to the people that matter only to our business.
โIt has allowed our clients to determine, โWell, Iโm spending $4,000 on DL cards or offline marketing, the traditional things that real estate agents do… [or] Iโm spending $1,000 over here on Facebook.โ The difference between the two is that when itโs digital itโs far more measurable and you can determine, โIs my money better spent over here?โ We would never suggest โcancel this one for this oneโ. But you can make a much more educated decision on where you should spend your money.โ
Kylie Davis of CoreLogic, facilitating the panel, also acknowledges this. โUnder no circumstances can you have a conversation that says, โI know all these things about youโ.
โYou canโt have people ringing prospects and saying, โour data tells me you are about to sellโ. So now we need to have a conversation about how to manage the slightly ‘creepy’ side of big data and how to โde-creepโ it.
“Itโs more about providing the right information that โmagicallyโ turns up at the right time or when youโre in the right headspace for it,” says Davis.
Over the last few years, notes Cobb, it has been harder to reach people on Facebook, no matter what industry you are in, unless you now pay Facebook to promote your post. In fact, Cobb has been working directly with his real estate clients with the CoreLogic and Quantium QSegments and says the results have been outstanding in both cost reductions and conversions (see breakout ‘Letterbox dropping on steroids’).
A. DIY CONTENT MARKETING AS AN APPROACH
Like any marketing, whether itโs advertising or content marketing, it comes down to the creative and how appealing that happens to be. But real estate businesses can be smarter at creating the right message when the audience is more defined.
Content marketing has been a buzzword in the real estate industry of late and is starting to resonate as a way to build an audience, but not necessarily target specific listings. Content marketing is more about providing value to your local market so that when the time comes to make a decision and list a property, your audience knows you and then may not consider any other options.
Says Brooke, โIf a real estate agent is advertising on Facebook with a message that just says, โList your house nowโ, it doesnโt appeal to anyone. If youโve got a picture of a lovely house in the โburbs with a young family holding up two kids [and that happens to be your audience] the mind tells you, โThatโs me, theyโre talking to me. This resonates better with me [as a customer]. This is more relevantโ.โ
Says Cobb, โThere’s two types [of approaches], I think. [There is] when we’re running a campaign for a particular property, or a development site, for example. And then there’s content marketing. So blog articles, market update videos, content-driven ads.
โSo on the property side, it could be a development site, and we might be targeting those ads to a specific audience. If we know for a fact that most people that buy units or off-the-plan apartments in a certain area come from within a 10 kilometre radius, and are investors themselves, then we might use Facebook ads to target exactly those people.
โIf it’s content, specifically, it might be an article about renovating that we’ve just written and uploaded to our blog, on our website. So we might target people who are more likely to be interested in articles about renovating.
“So I think there’s two types of ads, or promotions we can run on Facebook; either property or content-related.”
However, as with any marketing efforts including those using big data tools, you still need to define what your content marketing objective is, and how you will measure success.
How Can We Better Use the Data the Industry Creates?
APART FROM THE DATA THAT CORELOGIC and our big data panellists create and analyse, there is a huge amount of data that the industry itself creates. But there are some issues in making the practical stuff happen.
Brooke believes there is a wide gap that could be closed if data was to flow seamlessly between real estate franchise offices and head office, something that is being overlooked by many real estate brands. In other cases, information is being held up because each party believes that the data belongs to them. So while some are successfully sharing data, he says, most businesses would be amazed at how much crossover there is right now between buyers, sellers, tenants and landlords. Three or four differentfranchisees might be speaking with the same person, which really no longer needs to happen.
โWeโve just gone through the process of selling and buying, and Iโm being contacted through three different offices from the same franchise,โ says Brooke. Greater Data is working with some of the larger franchises to bring their data together with more intelligence and cohesion.
โFor various reasons, and again I do understand them, there tends to be a bit of a protection over the data and of people’s customers. The biggest problem that causes is, as a group, you won’t understand who your repeat customers are. We’ve seen it. We’ve seen it with groups where individuals have purchased from two or three different branches. ”
โWithout centralising that data, you will not be able to understand who those repeat customers are. By understanding who your repeat customers are, you can help sell them more products moving forward. For example, if you can spot somebody who’s bought an investment property from two different branches, guess what? Go and speak to them from the third branch about more investment property,โ says Brooke.
Says Cobb, โThe number one goal for driving traffic to your website, to engage with your content, in my opinion, should be gaining email subscribers. People who read articles aren’t necessarily interested in selling or buying right now. And I think that a big mistake that a lot of real estate professionals make is assuming that someone who reads your article, as soon as you put a call to action at the bottom, is they are ready to list, or need a property manager right now. It’s not necessarily the case.โ
Cobb also says that the goal of converting email subscribers is to continue the conversation with them until they are ready. โBecause we’re establishing trust, and by putting a call to action at the bottom, ‘List with me now,’ I think you start to negatively affect the ability to continue that conversation. So, instead of asking for the business right away, just ask for their email address and get them to keep getting that valuable content from you, ongoing.โ
Greater Data are developing a range of end-to-end marketing solutions, from targeting all the way through to marketing enablement and measurement. โWe are working with a couple of agencies to offer full packages to real estate groups, that perhaps donโt have a large marketing department and donโt have the resources to execute on some of the more progressive channels.โ
B. THE END-TO-END OUTSOURCE APPROACH
Brooke continues, โIn real terms, what it means is we’ll be working with real estate agents to first of all clean up their existing data sets; it’s very, very hard to use somebody else’s data effectively if you don’t understand your own data. You need to know what your data is telling you before you can then replicate and extrapolate that into other data sources.
โThe second thing is, once we understand who a group’s existing customers are, we help them to enable marketing to those consumers. Increasing brand advocacy, increasing sales through the existing customer set. As we know, it’s five or six times more expensive to find a new customer, than to resell to an existing customer. Our products are very much geared towards that.โ
3. WHAT ARE THE IMMEDIATE CHALLENGES FOR THE INDUSTRY?
Macquarie’s most recent โpulse checkโ of the industry concluded that an agencyโs database is the most important source of business, accounting for around 54 per cent of new listings. It also states that, โThe results suggest that agencies should ensure the foundation skills of prospecting, database management and lead generation become a key focus.โ
In a broader context, how can agencies take advantage of the databases that could be at their fingertips?
A. SHARING AND KNOWLEDGE COULD BOTH BE IMPROVED
Many of our panellists expressed concerns during the conversation about the current state of the industry in both training on big data and also storage of existing data. โIf databases are still held in Excel at franchisee level, if we donโt have the ability to understand existing data, we canโt, with confidence, step into big data,โ says Brooke. โTo understand acquisition properly you really need to understand what existing customers look like.โ Driscoll explains it in terms of building digital rapport. โIf we really simplify it, an agent would go into a listing presentation, and a good agent would always be looking for triggers. They might be looking for the Sydney Swans jersey on the wall, or they might be seeing a couple of [photos of] children on the mantelpiece who go to a particular school; whatever it may be.
โThen [the good agent] will talk about that and build rapport. I think essentially what big data enables you to do is to have a conversation, digitally or otherwise, on a relevant level to that consumer. You can’t expect [consumers] to fit into what we want; we need to fit into what they want.โ
Driscoll also believes the industry needs training to equip its participants with the skills it needs to understand the opportunities and the potential. โOne of the major issues is to do with upskilling. Weโre still rolling out the same kind of training as an industry that we were 20 years ago. This isnโt futuristic stuff; this is stuff people in other industries are doing now.โ
B. GETTING EVEN SMARTER BY KEEPING THE LOOP GOING
The big data loop also does not stop with simply getting your data in order. Itโs about having a continual feedback loop and learning more from your experiences to benefit the industry as a whole.
Do you know how many calls or visits it takes to list a property? According to Mitchell Prevett of Quantium, this is one of the simple yet critical questions that big data needs to answer so that agents (and BDMs in property managmeent) can refine their marketing and communication plans with potential customers to save time and eliminate unnecessary waste. He says itโs the basics that you need to get right first.
โI think something like two-thirds of data scientistsโ time is spent on cleaning and repairing data, then building the algorithms and applying them.
โAll the time we see different BDMs in different industries talking about their data, but they donโt capture things like how many times they visit clients, how many times they call them; itโs really simple information like that can which can make a huge difference when youโre going back to try to target them.โ
The key, he says, is to get your data clean and then intelligence can be applied over the top. User applications and other smarts can grow from there.
Says John Owed from Data Republic, โIf youโre using SmartList, which is derived from big dataโฆ when you capture a response, it enables a whole new set of data to be collected. So what is the message that you delivered? What response did you get back? Then feed that back into the algorithm. The amazing thing about big data is you are not just limited to one point in time; the data you collect has a network effect that you can keep building. Itโs a continuous cycle, and itโs phenomenal.โ
4. THE CONSUMER’S VIEW AND THE FUTURE
A quick summary of the things discussed in this round table that are available now:
1. You can predict, with some certainty, which house in a street will list next
2. Tailor marketing to that consumer to appeal to them at their stage of life or circumstances
3. Reach consumers in other areas that have a similar profile who might want to purchase in your area.
In the future, what will big data solutions look like to the consumer? Driscoll says it might be that the consumer will have the ability to determine their suitability for a house by themself. โ[I think] there will be some sort of artificial intelligence or algorithm to determine how suitable either my family or I am to a particular house. I think surely this isnโt that far off.โ
Then the question becomes, if youโre an agent selling the virtues of a house and the ‘algorithm says no’, where does that leave everyone?
Owed is still of the opinion that the ‘listing or buying’ targeting discussion will continue to be only part of the equation. โFrom a real estate agentโs point of view, I think they are always going to be involved because it’s such a huge purchase; people need to talk to someone.
“[But] there are so many things that you could build into a ‘productionised’ system, such as recommendations around how they should deal with people at each step through the process.โ
For Davis, itโs about human connection โ as well as perhaps stating the obvious. โI guess big data is about human connection and building relationships. What big data is doing is allowing you to amplify your ability to connect with more people, more quickly and more affordably. If you’re not great at [connection] now, big data is not going make you any better at it.โ
Then there is the marketing side of things. Says Driscoll, โI certainly think CRM systems themselves in Australia will develop and advance over time to the point where youโre starting to put the information into a CRM, and they will have interfaces or some connecting portal as a conduit to this type of information. You wonโt have to go to a third party; youโll be integrated. All of a sudden you are entering someone’s information [into the system] and there is a match.โ
Brooke summarises todayโs discussion in terms of agents being more in control over their destiny. โThink in terms of driving your own marketing, not just throwing something at a portal, crossing your fingers and hoping you get people interested. You’re actually testing what you are doing, youโre learning, you’re using all of the data available to you; learning to bake your own bread almost.โ
But can big data and social media take on the traditional portals? Says Driscoll, โ[I think now] Australia has the highest usage of social media, per capita, anywhere in the world. One of the things I’m sure we’ll see emerge is more properties actually being listed and sold, through or via social media. One of the things that we’ve just launched, albeit in its trial stage, is we have the ability to list or advertise properties natively within Facebook.
โWe’ve done some fascinating research in there,โ continues Driscoll, โNow the first thing is, we own that data information โ unlike if we advertise the same property through the portals. The other thing is that we can look at the analytics behind the scenes, so we can figure out if we are targeting the right people.โ
And the numbers seem encouraging too. โIf we can start leveraging some of this information and can work to our advantage…in our trial, two weeks ago, we took just the one property, and we looked at it across the portals and across native promotion through Facebook. Through Facebook, it was a quarter of the price, and we had pretty much the same inquiry level.โ
5. ACTIONABLE INSIGHTS
Here are five actions that the panel thought were relevant to all agents right now:
1. Be curious about what others are doing around you; what is working, whatโs not. Learn how the technology works and consider how you can use it to save time and money.
2. Start with your data and get that cleaned up and under control.
3. Measure and learn what you are testing and trialling. Use content marketing with messages that appeal to your target audience.
4. Donโt write everything off if you have a hiccup or two; you need to be consistent. Remember, itโs a prediction and not a cast-iron guarantee. Donโt be afraid to engage the experts; thatโs what they are there for.
5. Donโt be afraid of sharing data. The better the data that underpins the analytics, the better the result will be for everyone.