Real estate's real data issue

Target knows when I’ll get pregnant. HP knows when I’m about to quit my job. Pandora knows what new music I’d like to hear.

Companies are mining data to predict what I’ll do and when. As I’m sure you’ve heard, “Big Data” is kind of a thing.

Yet we haven’t seen much of this sort of data mining and predictive analysis happen in real estate. But we should, and soon.

In an industry where the majority of consumers only participate once every 5-10 years, and brokerage brand loyalty is weak, knowing things about a consumer’s life trajectory is more than just handy information. It could be lifeblood.

Imagine you know when your past clients are ready to sell or refinance. You know when their kids will be buying their first homes. When they may look to downsize or buy a little pied-a-terre. When they’ll be struck by the vacation home bug.

Imagine you know just the right moment to send them an informational email about relocating to another state. Or when to hit them up with enticing photos of bigger homes in their area because they’re expecting another child.

Creepy? Maybe. But indeed powerful.

After all, great marketing is just as much about timing as it is about content.

This may sound like idealistic dream talk, but this very sort of predictive targeting is already happening. Look at Facebook’s announcement this year that they would begin enabling advertisers to target consumers on their platform based on those consumers’ offline shopping behavior.

One company that we know of, SmartZip, is trying to analyze and predict consumer behavior for real estate. But it’s just the tip of the iceberg.

For a real estate company, leveraging predictive analytics in this sort of way requires a long-term strategy. We’re talking years here, which sounds painful. But hear me out.

You’re probably already collecting a fair amount of data from your website –  email at the very least, and search habits for the more advanced. You may even be using drip campaigns right now to try to keep a connection going with past clients and prospects.

But do those drip campaigns really work? In most cases, no. (Think back to the “turn your clocks back” message I brought up in my last post.)

You have to take your data collection and mining practices to the next level. Establish connections between business units (brokerage, mortgage, title, warranty) track that data, and map it to email campaigns that take place over several years.

One simple way to start would be to create an exit survey for all your clients (yes, if you’re a broker this is tough; if you’re an agent it’s a no-brainer) that captures some basic information about their potential future home-buying plans. Instead of pounding them with weekly emails about the market right after they bought their house, save it for when they’ll be more apt to pay attention to it.

Put all your recent buyers on a 1-year new homeowner campaign that speaks to them as if they’ve just bought a house (because they did!). Send them things about homestead exemptions, taxes, home warranties, refinancing, home improvement. Don’t send them new listings or “It’s a great time to buy” messages – they just bought a house, remember.

Another way to tackle would be to really use your local knowledge to make some predictions.

For example, where I live in Oakland, it’s common for people to move because of their children’s school choices. People often will move into a neighborhood that has a good elementary school, but you can expect they will be looking to move again when their child hits middle school age.

Bottom line

Your customer data is just as valuable as your listings data. Maybe even more so, if you can crack the code on how to match it to marketing messages and timing.

Behavioral data is already pushing other industries forward. Spray-and-pray no longer works in advertising or marketing. Now, it’s all about targeting and context. And the only way to do that effectively is to know more about the people you’re talking to.

This is the type of data that is pushing other industries forward. Are you game?