At one time or another, we’ve all experienced information overload as we’ve tried to sort through all the data saved on our computers, stacked on our desks and stored in our smartphones.
That same issue faces loss prevention specialists in the retail industry on a daily basis as they try to make sense of both historical and real-time events and activities captured in Point of Sale (PoS), store management, video, intrusion and access control systems.
This compilation is often referred to, appropriately enough, as Big Data: a heaping pile of bits and bytes that needs to be converted into useful information.
So how do you make sense of Big Data and put it to work for you? Think of it as you would the New York City phone book. If you were looking for someone named John, without having any other parameters with which to find him, you could spend days or even weeks going through all the men named John in the book, checking out each one. But if you know that John’s last name is Smith, you’ve applied a rule that narrowed down the list considerably. It’s still likely to be long, so you look for an opportunity to narrow it even further, and look for all the John Smiths on Lexington Avenue. Now you’ve set the threshold for a workable list.
The same concept of sorting takes place when you apply analytics rules to Big Data. Each rule you add can narrow the data pool until you get to a level where the information is usable and actionable. For example, if you start with the number of people who walk into a store each day, that number may be so large that it does not provide any actionable intelligence. But by applying rules — how many of the total number of people in the store actually made a purchase — you can get to a measurable result. So, if 700 people entered a store on a Sunday, but only ten made a purchase, and only between noon and 1 p.m., you have a result you can address.
Big Data can also be culled to address security issues such as employee theft. You know from applying rules to all your data that you have 100 returns to your store each day. And by analyzing the data even further, you find out that of all the registers in your store, 20 returns are from a particular unit. That raises a red flag. By adding more rules using your video surveillance data, you look at all registers to see when there are returns with no customer present. Now, with this narrowed list, you can identify your likely source of employee theft. Advanced tools provide you with the ability to execute a search from two isolated databases and narrow the results based on the match of the combined data.
By identifying and deploying specific applications to Big Data, you can achieve the goals you want for your business, whether it’s related to improving security, spotting business trends or gathering some other form of usable information.
The key with Big Data is to keep applying rules until you end up with usable, measurable, actionable information. From there, the sky is the limit as to what you’re able to find among all the bits and bytes you’ve stored up over time.
If you’re interested in learning more about the power of data mining and the usability of Big Data, please click here to view our recorded webinar, “Big Data: How to Combat Crime, Spot Business Trends and Determine Real-Time Traffic.”