Using Big Data for Retail
Using Big Data for Retail
Let's see how Big Data methods can be adapted for brick and mortar stores by focusing on how much we can answer the 5 fundamental questions of retail through Big Data.
Join the DZone community and get the full member experience.Join For Free
Access NoSQL and Big Data through SQL using standard drivers (ODBC, JDBC, ADO.NET). Free Download
As the twenty-first century ushers in new advancements, many expect a decline of brick and mortar (B&M) stores while their online counterparts flourish. However, not even Generation Z, those born in the digital era, are ready to give up shopping in a real location. Although online stores usually offer a better price and vast product diversity, the sensorial experience provided by a store is hard to match. B&M locations are here to stay since they offer the instant gratification of owning the product, personal services with a human connection, and an immersive space. Some buyers do not even think about taking out their credit card until they have tried the product. They use stores as showrooms, a trend introduced by Apple that will continue to grow.
Big Data for B&M
Until now, the B&M strategy has relied on promotions, sales, and convenience to increase revenue, but they are starting to learn from their online counterparts. In a digital store, there are about a dozen factors to be measured and recorded about the client's interaction. In a real world setting, there's a lot more recordable information that can be used to increase profit. The recordings include product choices, walking paths, time spent in the store looking at items, a person’s mood, and interaction with sales assistants. Using Big Data to record and correlate all of these with the sales record to determine preferences can be much more efficient than the current uses of Big Data for online shopping.
Let's see how Big Data methods can be adapted for B&M stores, as a bit more engineering is required in collecting the data points. We will focus on the five fundamental questions of retail and see how we can answer each through big data.
Who Is the Client?
Where online stores usually stop at county-level or city-level and some simple demographics, B&M can go in depth and collect much more relevant data through CCTV and even personal interaction. Facial recognition technologies can approximate age, race, and even mood. Transforming people into data points shows their habits like shopping alone or with their families. Adding Wi-Fi and Bluetooth technologies can enhance the recognition of returning visitors and setting in motion appropriate greetings and discounts.
Learning who enters the store can help a business direct their targeting efforts better. This means moving away from guessing the most likely buyer and moving toward knowing them on a personal level and being ready to serve them.
What Are They Doing?
What are potential clients interested in? Which products are they holding in their hands or trying on? What makes them buy or abandon the intention? Combining security camera footage with RFID (radio frequency identification) tags on the products can create a matrix of favorite products and a list of possible reasons for not buying.
Combining in-store info with the cookies from a client's browsing history and social media profile scanning creates a complete user experience. This can be the first step in customizing buying offline in a similar fashion to how the online shops do it. Most customers may not find these practices intrusive since they are already accustomed to the service. Even more, if one B&M store starts to offer this, they would set the bar high for the competition.
Where Are They Going?
One of the main differences between digital and B&M stores is the possibility of tracking walking paths, creating heat maps of locations, and increasing staffing in busy sectors. The video footage already available can track every individual as a point on a 2D map of the store. This will indicate the most relevant products and displays as the focus points. These products are the ones worth being promoted, both online and offline. This practice is like measuring time on page, but offers more insight since behind the browser window, there is no tracking of what the customer is doing.
Speaking of the online-offline continuum, the lines are becoming blurred. Shoppers check prices online while in-store. Half of them would spend more time in a physical location if they could browse the inventory on the computer, a process called web-rooming, the opposite of showrooming (checking products in store and buying online).
When Do They Shop?
Most physical stores have a ten- to 12-hour window for their opening hours, and within this time-frame, they experience only a few peak moments of high traffic. Using Big Data to track these occurrences helps managers create better staffing patterns. This tracking can go up to an individual level by using beacons for proximity marketing. When a client is approaching the store, you can trigger a push message with an offer, luring them to come in.
Analyzing paths and correlating them with the time can create maps that show the busiest sectors of the store by the hour. This will assist in creating a dynamic promotion pattern. Think of it as a Big Data enhancement of the traditional morning or evening happy hours.
How Can We Comply to Their Needs?
As described, using Big Data for B&M stores takes more than just recording information and analyzing it. It is a domain at the border of statistics, engineering, and behavioral psychology.
Shops that have decided to follow this path should be prepared to invest in technology. From simple mobile POS and tablets for shop assistants to VR-enhanced fitting rooms, high-tech is finding its way into the retail world. Throw in some guest Wi-Fi recognition and never cease to A/B test new ways of approaching the client. Also be prepared to have your store fitted with interactive screens or other gadgets that allow the client to communicate with the system.
Big Data is no longer just an online tool. It is quickly becoming an offline retail enabler. Since each trip to a store can generate over 10,000 data points, it would be foolish not to harness the power of such customer insights. It is now just a matter of time until this approach will be the norm in retail. The real challenge now is to find suitable tools to perform storage, cleaning, and governance; and integrating Big Data in existing systems is another requirement.
Opinions expressed by DZone contributors are their own.