What makes big data big? And why is this idea of ‘big data’ more valuable than the data which is sitting inside the database of an ordinary real estate agent?
Here are the nine things that determine how valuable your data is – or isn’t – ranked from lowest in value to highest.
1. Static Data
Most agents have static data – information that is keyed into a system at a particular time and which may never again be updated. Static data is helpful – it’s a record of what you’ve done or who you know and what happened which can help you see patterns of the past. But it’s just the start of the big data adventure.
2. The source of your data
How was your data sourced? Most agents collect their own data at open for inspections and cold calling. They add this to sources of data such as the data within RP Data Professional. The value of ‘source’ is based on the integrity of the collector. If you know that the person or organisation collecting the data has high standards of accuracy, the source of your data can be an asset that is important to its value. But if your source is a bunch of agents who are only typing names into the database because they have to, the ‘source’ may be a liability.
Cleanliness is next to godliness in databases, and like housework, requires constant vigilance and maintenance. If Mickey Mouse and Donald Duck are listed as potential buyers in your CRM, you know you have a problem. Identify key data points that must exist in your database (such as full name, address, mobile and email) and make these your minimums. Then, set aside CRM housecleaning so that records that don’t have these minimums are contacted and either corrected or culled.
While obviously, big data requires scale, bigger is not necessarily better in a local real estate agents database, if that size comes at the cost of the first three criteria. A big database that is dirty, inconsistent and lost in time is not as valuable as a small, well maintained and sourced set. Get the first three right though, and adding size does add to the value.
Did you know that data is like milk and has a use-by date? Leave data too late, and it goes off, the information becomes too dated to be valuable. Being able to have your data in a state that allows it to be accessed and used quickly can be more valuable in some circumstances than having a dataset that is perfect but outdated.
What kind of insights does your data hold? What are the similarities between your customers and how they transacted? What are their ages, life stages, expectations? How long is it taking them to move through a property cycle? What does a typical buying or selling cycle look like? How is your business accommodating these similarities and differences?
Many people shy away from ‘examining’ data thinking you need a tertiary degree, but sometimes, just making time to look through your data on a regular basis and asking yourself “what’s it showing me?” or even “what’s changed?” can provide you with valuable insights.
Insights however, are only truly valuable if you know what to do with them. So a database that helps you both see the insights, and then helps you follow up those insights with action has a higher value.
What constitutes an actionable insight? One where you see that time on market is starting to rise, which enables you to double down on selling the value of marketing campaigns to vendors. Or one where you notice that you’re attracting more women in their 30s as open for inspection attendees whenever you promote a property on Facebook, and so decide to increase your Facebook promotions and improve their wording and targeting.
We saw before that data can be static and dated and still have some value. Timeliness is the complete opposite – data that is not only fluid and fresh, but accurate, clean, significant in size, insightful, sourced legitimately and enabling of actions that allow you to respond and make decisions quickly.
Significant computing power and analytical smarts are required to create timely data, but the rewards are dramatic. In stockmarkets, computer algorithms crunch share information to allow trades to occur tenths of a second faster than competitors, such is the value of the arbitrage. Real estate data is heading in a similar direction but has a long way to go.
Once your data has passed the timeliness benchmark, the final value layer is its ability to be predictive.
When you know everything there is to know about your past customers, what they look like, how they behave, typical transactions and what their value is, there are two significant things you can do with it.
First, by seeing how past customers have behaved under certain circumstances, you can start to apply that knowledge to new customers, pre-empt problems and deliver a better standard of service.
Secondly, you can apply your customer insights to the rest of the known data universe to identify people who look exactly like them with whom you have not yet done business. Facebook call this ‘look alike audiences’ but Facebook is not the only way to find lookalikes.
In this way, if you know that your ideal past vendors were 30-something women with two kids in a three bedroom house, you can target other women in their 30s with kids to become your next set of vendors.