Audiences in real estate & development are craving insights to help them better cater to the shifting needs of consumers. The key question they are hoping to answer: What drives the purchase decisions of the people who are ready to move? Being able to capture this information can help at various levels – from cost efficiency at the design & build level to developing more accurate marketing campaigns.
In this guest blog, consumer behavior speaker David Allison shares his data-driven insights for real estate event attendees to use to their immediate competitive advantage.
Who Are We Really Building This For?
“…the real shocker was that everyone also agreed they’d pay more (as much as 15% more, on average) to live in a building with people who cared about the same things; people who shared their values, regardless of how much they resembled each other from a demographic point of view.” David Allison
I spent more than a decade as the founder and president of a marketing creative and strategy firm that specialized in real estate development. We helped create brands and campaigns for residential, commercial and recreational projects all over North America and beyond. It’s been almost four years since I sold the firm, and while the rate of change in the industry has been aggressive, at least one aspect of the business remains the same. Ironically, it’s perhaps the most flawed component of the process, the one thing that should be changed to create the most dramatic improvements.
It always started the same way.
When I sat in those boardrooms for all those years, there was always someone full of certainty about who was going to show up and buy or rent the suites and therefore how the building or community in question should be designed and built. They often expressed their ideas about the target audience in the form of a demographic description, and occasionally had gone as far as writing up a profile of a typical prospect.
You know the document I’m talking about…it would include things like: “Sally and Bob are baby boomers who are looking to start a new life in an urban and walkable neighborhood. They had three kids who have all left home now, and they want a two-bedroom floor plan so that the kids will have a place to stay if they come to visit. They both drive upscale vehicles, and Sally wants the spare bedroom as a place to keep her scrapbooking supplies when it’s not being used for overnight guests.” And it would carry on from there, telling a nice story about this typical couple who were going to snap up all the homes.
But the most peculiar thing kept happening.
Years later, after the building had sold or the suites had been rented out, we’d look around and see who had actually decided to live there. And while there may have been an occasional ‘Bob and Sally’ they were only there by chance. The building had sold to all kinds of people who most often had no apparent demographic similarity to each other at all. An entirely different audience had ended up buying than the one that had been targeted. Looked at another way, an entire building or community had been built for people who never showed up. It happened over and over again, but we all more or less shrugged it off, because after all the building had been a success and it was fully sold or rented. So what did it matter? And we’d all go off and repeat the process on some other building somewhere else.
What if the people we thought were coming had actually shown up?
What if there had been accurate predictive insight behind the target audience profile, instead of well-meant guesswork and opinions about Sally and Bob? Not only would the risk of every decision about the design and marketing of the building been minimized, but we’d have been confident and bold. We’d have created more specialized product that the occupants truly loved. In turn that love would have rubbed off and accrued on the development company brand, making every project that followed easier to sell or rent.
How can we make this happen?
Sociology is the science of understanding the behavior of groups of people, and if our goal is to more successfully predict the behaviour of target audiences, we need science on our side.
Ask any first-year sociology student what makes people decide to do things and you will get the same answer: what we value determines what we do. If you value family more than anything else in life, you will make every decision (consciously or subconsciously) based on how that decision will impact your family. If we could detect what people care about more than anything else in life – the values – of a target audience for a development project, we could predict their behaviour.
There’s been a major obstacle to doing this. In order to accurately profile the shared values of an entire target audience, we’d need an enormous amount of data about values from all kinds of people. Until recently, data collection technologies that allowed for a complex dataset like that didn’t exist.
Data collection in the real estate sector has always been about two things.
First, we are really good at analyzing past sales of similar projects, or past-purchasers from a sales database. That’s a great way of finding out how people behaved yesterday, but not really a great way to predict how they will behave in the future.
Second, we occasionally send out surveys by email or have a few focus groups with people who match the “Sally and Bob” demographic description.
But those methods mean developers are placing multi-hundred-million-dollar bets based on the opinions of a handful of people who might share similar birthdays or income brackets. Or people who bought real estate before. It doesn’t help us detect the values of the people who will buy today, and understand how they will behave.
Algorithms to the rescue.
Thanks to algorithmic data collection and analysis techniques we can now amass enormous datasets that were unimaginable only a few years ago.
For the last three years, working with a university research team, we’ve assembled 100,000 surveys in a format that statisticians refer to as a random stratified statistically representative sample. Simply put, it is an exact replica of the real world, in miniature, with the same proportionate number of people of all ages, incomes, education levels, marital status, geographic locations, and so on. The dataset contains 380 metrics about 40 core human values, and which values are most important to which people, and why.
The data is +/- 3.5% accurate with a 95% level of confidence, which means we have a redundantly redundant amount of data, and it’s all more accurate than you’d need for a Ph.D. thesis at an Ivy League university.
To start with, we found two fascinating things.
First, we discovered proof that demographic profiles are not effective.
Across all those surveys, baby boomers disagree on everything 87% of the time, and millennials disagree 85% of the time. How can we make decisions about targeting groups of people who disagree so wildly?
Those numbers indicate that for every dollar or hour you spend targeting boomers or millennials you will have about a 13% – 15% chance of triggering a shared value, and therefore influencing behaviour. On further examination of the data, using ANY traditional demographic segment, you are going to get those same dismal results.
We now have proof that demographics, and the stereotypes attached to them, are only negligibly effective. Millennials don’t all like avocado toast. Boys don’t all like blue. Rich people aren’t all that different from the rest of us.
Don’t get me wrong, we still need to define our target audiences using demographics. 18-year olds are not going to buy three-million dollar penthouses after all. Using demographic stereotypes about those penthouse buyers to determine what they want, how to motivate them, and how they will behave? The stats show just how dangerous that can be.
Second, we verified that the sociologists were right: values are an extremely powerful way to profile and predict the behavior of a target audience.
Values-based groups agree with each other across all those 100,000 surveys and 380 metrics as much as 89% of the time. That’s a factor of seven or eight times more agreement, more similarity, more predictability for a target audience than demographic profiles provide.
What did we learn about real estate developments?
The data was unmistakably clear on this point: no one wants to live in a building full of people who match their demographics. Across all ages and incomes, everyone agreed that monoculture buildings or communities full of people who were demographically the same sounded like an undesirable place to live.
But the real shocker was that everyone also agreed they’d pay more (as much as 15% more, on average) to live in a building with people who cared about the same things; people who shared their values, regardless of how much they resembled each other from a demographic point of view.
Let’s combine some of these ideas.
Start with this: if we profile a target audience using shared values (which we can now layer on top of a demographic description for a building or community) we can more easily and boldly design and market homes that people will flock to, and love, and only leave reluctantly.
Then add this: if we do that work properly, and build homes that reflect the values of our target audience, they will pay more than market value to rent or own there. We will need to make it obvious that their values have been considered, and clear that they will be living in a vertical village filled with people who are there for the same reasons.
Which leads to this: we can build buildings that will make people happier, and make more money too. Simply by changing how we look at the people in our target audience.
So it turns out the old saying is true.
Birds of a feather DO flock together. But not because of how old they are, or how much they resemble each other on the outside. It’s what’s inside that counts.