Learn the difference between good, bad, smart and dumb data, and how to use it.
Like many industries, the multifamily industry has embraced and developed technologies to make our decision-making processes more efficient and our businesses more profitable. At the very core of all of this technology is information – data. No matter where you are collecting your data from (whether it be property management software, utility billing providers, utility companies, financial institutions, third-party information technology providers, vendors, Google, etc.), data is available to us in quantities, and qualities, unlike ever before.
So much data…So now what?
Well…is it any good?
To simplify how to qualify data, here are two questions to consider:
1. Is this good data or bad data?
2. Is this smart data or dumb data?
The first question seems simple, but the answer can be ridiculously complex.
Good Data or Bad Data?
Obviously, we don’t want to make any decisions based on bad data. But how do you know if it is bad data? The scary thing is, you often don’t. Start thinking about where this data is coming from, why this is data being used this way and how does this data meet your objectives. Let your supplier partner explain themselves. It’s not enough to hope you are not using bad data. For whomever is providing your data, even if it’s your own data, these systems need to have measures in place to protect your decision-making process from bad data before it is ever received. Warning, most supplier partners and software systems don’t have meaningful protections. Bad data is frightening because it is hard to locate and hard to see.
Here is an example: Most property management software systems will let you post a single rent payment more than once within the same period. So instead of having a $950 rent payment properly credited to a resident’s rent obligation, many software systems will allow different employees to post the same rent payment, resulting in a $950 credit due resident. This same credit is picked up by a property’s utility billing company and next month this resident can march into a leasing office demanding his money or refusing to pay his next month’s rent and he has the proof in his hand, his monthly convergent bill. There are a lot of typical processes in this example and yet bad data sailed right into your resident’s hand. All because the property management software did not have a failsafe to prevent this.
Smart Data or Dumb Data?
Dumb data isn’t dumb in the sense that it is stupid, it simply means that this data is not dynamic, it is only a reflection of numbers and values. Dumb data is often found in the form of data files or spreadsheets. A perfect example of dumb data is your rent roll report, a line by line report of every apartment home, resident, lease date and recurring charge on your property. You look at your rent roll report every day. They are full of information and can be used as a great tool. However, if you have a 400-unit property and there is an error in your property management software, good luck finding that one error amongst 400 lines and a dozen columns on your rent roll report, for each property in your portfolio.
Smart data can take many forms but you always start with your goal and what data is important to the decision process. Smart data considers only what is important and is designed to mine through large amounts of data to drill down to specific answers to questions you may have. Smart data can weed through those 400 lines in your rent roll report and find the unit that has an incorrect rent set-up. Smart data finds and protects you against bad data.
Here is another example: A 300-unit property’s August water bill is $13,500. Last February, this property had a $12,000 water bill. A typical analysis of this $1,500 increase over six months could warn of a 40 percent increase in the property’s water cost over the calendar year. The data is good data, the numbers are accurate, but the analysis is wrong. This data is dumb data and doesn’t consider both what we want to know nor does it consider what is important. In this example, what is important is that the winter, February water bill was $12,000 for a 28-day service period while property occupancy was 92 percent. August’s $13,500 summer water bill was for a 33-day service period and occupancy improved to 95 percent. Smart data that considers what is important in our example would make the decision process very clear. Instead of a 20 percent, six months increase and a seemingly 40 percent annual increase in water costs, it turns out that our property’s August water bill is 7.5 percent lower than February’s bill and seasonally adjusted is 14.4 percent lower.
Starting with good data and then considering what you want to know to create smart data, can help you make educated and powerful decisions. Demand more from your supplier partners, your software and billing providers. It should be each party’s responsibility to safeguard from bad data and to enhance our decision-making processes with smart data.