It is an industry characterised by diverse business models, complicated products and contracts that span decades. And it is the complex nature of the real estate industry that makes the use of business intelligence technology within the field so difficult and at the same time so valuable, says Chris Lees, executive director of Calvis, an IT and management consultancy that operates in the professional services industries.
“The real estate industry is enormously diverse,” explains Lees. “Even within a single organisation, there might be a huge diversity of clients, contracts and revenue streams.”
That diversity is matched by a complex distribution of data, he adds. Most of the information that a typical real estate management firm receives will be in an unstructured form, such as emails or word processor documents. What structured data there is can be found in departmental applications, self-built systems and, overwhelmingly, multiple spreadsheets.
The scope of the data collected by real estate firms can also vary wildly. A firm might need to keep details of a long-term lease for 25 years but it might also need to track security access card data on an hourly basis.
Something as simple as the address field in a customer record can be maddeningly complex in the real estate industry. A property might have six or seven different addresses associated with it – a grid reference, a planning reference, a postcode and so on; or, when the property in question is a stretch of sea bed, it might have none at all.
So why are real estate management firms increasingly interested in overcoming this complexity to build business intelligence capabilities?
“Within the complex mix of services that real estate companies offer,” Lees explains, “there are many opportunities for intelligent use of data.”
Those companies that overcome the data management challenge facing the industry can achieve more efficient property management and more sophisticated client relationship management, he adds.
Data in context
The broad variety of data architectures and business models in the industry has inevitably led to a broad variety of solutions, some of which challenge the observed wisdom of business intelligence.
One Calvis client was faced with the problem that the firm’s executives spent most of their time in meetings with clients outside the office. The business intelligence challenge was to deliver property portfolio information to mobile devices in such a way that it could be sliced and diced remotely.
As some of this data came from external sources, the company chose to integrate the various data sources using an enterprise service bus (ESB) on top of which a presentation layer was built, which pushed the necessary reports out to advisers in the field – an approach not commonly associated with BI projects.
There have, however, been moves in the industry to establish standard data definitions. Lees himself has been involved in the development of PISCES, an XML schema that describes standard objects relating to the property industry.
Industry-specific standards are a real boon in projects such as business intelligence, says Lees: “If your industry doesn’t have one, I suggest you get together with your peers and make one.”
But he adds that data standardisation itself “is fraught with danger”. Data without context is just data, not information, so to enforce standard definitions of data points across an organisation at the expense of context is a destructive impulse.
“After all, when people talk about business intelligence,” he says, “what they mean is placing data in a business context.”
Lees’s insights into the real estate world suggest that approaching BI from an industry-specific viewpoint may be one way to build the necessary context into BI systems.