Data Science Techniques for Real Estate
As discussed in RealWorld 2018 sessions
Most people won’t argue with you about the importance of data. There are more data sources now than ever before, and we know we need to tap into them—but collecting the right metrics, interpreting properly and translating into business decisions that improve results is another story.
Savvy property management professionals are bridging this gap by stepping into the role of the data scientist, no lab coats or beakers required. At RealWorld 2018, RealPage’s own data scientist Rich Hughes led data-driven experiments and put numbers to the test in “Data Science Techniques for Real Estate.” Hughes challenged attendees to rely on statistical evidence, rather than conjecture, when evaluating real estate decisions.
The proof is in the numbers
We can speculate all day, but numbers have power and convey truth. How are you evaluating your performance data? How well do you understand your market conditions and competition? Are you getting the right types of leads? The answers to each of these questions can be uncovered using data science.
As any good scientist knows, a thorough process needs to be followed when performing an experiment. Hughes recommends following the scientific method—hypothesizing, testing, analyzing and revising— when evaluating the key drivers of performance in multifamily.
Putting data to the test
Hughes led hands-on experiments during the session, walking through several commonly-held conjectures in multifamily to debunk the myths. These included:
- “Mandatory Renters Insurance negatively impacts revenue performance”-Hughes shared insights into a RealPage study, which analyzed leasing data for 156 communities nationwide. Led by Hughes, the study compared the revenue performance of properties that mandate renters insurance against the performance of similar properties that did not in the same markets, sub-markets and ZIP codes. Based on the statistical evidence gathered, it was determined that mandatory renters insurance does not negatively impact a property management company’s level of profitability.
- “Women don’t like ground-floor apartments”-While many believe that women prefer to live on higher levels, recent studies have determined that this is incorrect. It was found that preference level differs depending on property and unit type, but women were less likely to select high-floor units in high-rise properties for both multi and single-tenant units.
- “Older people don’t like stairs”– It’s often assumed that elderly residents want to live in ground-floor units, but the data behind this is not statistically significant. Again, preferences in this group differed based on if a property was garden/mid-rise or high-rise.
Next, attendees had the opportunity to conduct experiments of their own. They put numbers to the test to determine if additional conjectures had merit or were incorrect assumptions, with the potential to damage profits if used in decision making. Some of these included, “People with roommates stay longer”, “unit size affects resident retention”, “website color helps sell apartments” and “residents like birthday cards.”
By testing hypotheses using data, multifamily professionals can identify the greatest opportunities for value in their businesses and refine processes in alignment with these areas. Hughes recommends retesting periodically, as market environments evolve on a continual basis.
Maximizing revenue with predictive analytics
When you make a significant investment, you want the greatest profit possible. Unfortunately, most people fail to see the full return on their real estate investments. In a complex market, it’s easy to over to underprice units, which results in missed opportunities. Multifamily professionals need better, faster decision making to reach ideal performance.
Now, industry professionals can measure lead quality based on statistical evidence. Using the right revenue management solution is a game changer in maximizing profitability, as it leverages predictive analytics to predict future behavior based on past performance. Multifamily professionals can increase revenue premiums and gain a consistent pricing methodology to see a performance boost between two and seven percent. They can also reduce the price negotiation burden often placed on staff and improve leasing flexibility for residents—all using data science.
“There is now a better way. Petabytes allow us to say: ‘Correlation is enough.’ We can stop looking for models. We can analyze the data without hypotheses about what it might show. We can throw the numbers into the biggest computing clusters the world has ever seen and let statistical algorithms find patterns where science cannot.”- Chris Anderson, “The End of Theory: The Data Deluge Makes the Scientific Method Obsolete”, Wired Magazine
With actionable metrics, property management professionals gain an in-depth view of their performance. Learn more about how RealPage’s Asset Optimization solution provides deep data science, proven expertise and precision results.