Introduction To Applied Big Data For Stores

As the Big Data megatrend continues to skyrocket, we’ve seen the rapid development of its lesser-known cousin “Applied Big Data.” Applied Big Data is the practice of using new Big Data computing techniques, platforms, and tools to solve problems that would be unsolvable in the absence of massive quantities of data and the computing power necessary to handle it (versus the general, building-block platforms and components, like Hadoop). Now, we can get even more specific.

“Applied Big Data for brick-and-mortar retail” includes platforms that have the ability to measure and draw actionable inferences from customers’ behavior in brick-and-mortar stores in the same way that their online partners have been doing for years. Applied Data for brick-and-mortar retailers accounts for diverse factors such as the physical location of shoppers and employees in the store; the layout, fixtures, and planogram of the store; staffing schedules; complete detail on actual sales, and even the weather. Input sources can include video cameras, Wi-Fi tracking tags, smart shelf sensors, and other in-store systems like those for Point-of-Sale (POS), staffing, and task management.

Big Data = Even Bigger Results

The potential for retailers is as substantial as the data sets. Retailers can gain a precise, factual understanding of how shoppers move around their stores – where they go, in what order, how long they stay, when they come to the store, and how all of these questions map to actual sales. Retailers have optimized store layouts, fixtures, staffing, and even product offerings based on what they have learned. The upside for these companies is stunning. By taking an Applied Big Data approach to their brick-and-mortar locations, retailers have seen outcomes like these:


  • Montblanc increased same-store sales 20%
  • American Apparel increased same-store sales more than 30% and reduced theft 16%
  • American Apparel also combined its people counting and Loss Prevention platforms, saving 40% in capital expenditure for these two functions
  • Family Dollar remodeled more than 1300 stores in the first nine months of its deployment based on measuring and analyzing shopper behavior
  • Brookstone used analytics to reduce shrinkage more than 0.2%, improving the bottom line by roughly $1 million per year

Results like these are possible because the brick and mortar setting is one that traditionally has lacked such measurement. Instead brick-and-mortar retail has been forced to rely on blunt error-prone techniques like surveys, store visits, basic traffic counting, and context-free shopping basket analysis. In fact, McKinsey and Company estimates that Big Data analytics can improve retailers’ operating margin more than 60%.

4 Ways To Analyze Big Data For Store Optimization

Through these data sources, retailers using our system collect about 10,000 data points per store visitor. Each year we collect more than 50 petabytes (50,000,000 GB) of raw data across more than 300 million shoppers who visit our customers’ stores. We then process this deluge of raw data into trillions of analytical data points (one level more abstracted than raw data) and use them as the foundation of measurement and analysis to enable store optimization:

Performance Benchmarking

By calculating and comparing conversion rates for all stores, retailers can identify norms, trends, and outliers in conversion. This information is then usable in a number of ways. For example, you could discover your lowest performing stores and target them for training or support programs to bring conversion rates up. Or alternatively you could identify the highest performing stores and investigate them for best practices to deploy across the entire organization.

Staffing Optimization

Applying Big Data to brick and mortar stores can show you the traffic cyclicality you experience throughout the day, week, and year. Understanding your optimized customer-to-staff ratio enables you to provide services appropriate to your business model, ultimately contributing directly to maximum profitable conversion. In fact, by correcting misalignment between staffing level and traffic volume, stores often increase conversion without spending additional budget.

For example, Wi-Fi tagging makes it possible to track the movement of employees through stores. Retailers use this feature to remove employees from overall traffic counts and to compare employee movement throughout the store against customer traffic flow and against expectations for top performance.

In addition, retailers have used combined views of in-store movement and POS activity to understand and optimize cash/wrap staffing. Understanding queue times enables retailers to consider staffing needs separately for registers as opposed to the entire store, arriving at the best staff levels for each.

Marketing & Floor Layout Effectiveness
In the past, retailers could only measure the effectiveness of marketing campaigns based on actual sales results; in this process, a great deal of information that could have helped maximize marketing ROI was lost. By using Big Data analytics to understanding the path to purchase more completely, retailers now have deeper insight into why campaigns are effective or not, making more efficient marketing programs possible.

Big Data analytics can also reveal how much time customers spend at all parts of the show floor in a single, understandable image. Simply moving the highest margin products to more heavily trafficked areas can significantly increase sales and profits.  Furthermore, retail chains can easily test and tweak new campaigns, layouts and fixtures to ensure their best effectiveness prior to widespread rollout.

And combining detailed understanding of traffic and dwell metrics throughout the store can build a fuller understanding of store flow—each store’s unique traffic patterns by type, geography, weather, or other factors.

Fraud Detection & POS Exception Reporting

Loss prevention teams can view aggregated trends in POS usage to efficiently monitor for outlier behavior that may indicate fraud or theft. With Big Data analytics, identified POS events are viewable instantly.

Loss prevention teams can easily monitor employee compliance by identifying events in the point of sale system such as returns or voided transactions and accessing video of each event with a single click. Today you can even set up automatic exception reports to notify LP teams of unusual POS activity. And the store operations team can quickly find and view video of specific types of events (such as large purchases) to learn more about your customers. Each POS event can be linked to both a sales receipt and video, allowing one-click access to the full information behind each transaction.

As retailers become savvy to the benefits they can derive, we can expect Applied Big Data to sweep through and ultimately transform the brick-and-mortar retail industry, ultimately becoming a foundation of that industry just as measurement and analysis is a fundamental part of the e-commerce industry today.

Alexei Agratchev is Co-Founder and CEO of RetailNext, which provides real-time in-store monitoring and analytics. Prior to joining RetailNext, Agratchev was the founder and General Manager of an internal startup within the Cisco Emerging Technologies Group focused on developing video applications for the gaming and retail markets. During his eight years at Cisco, Agratchev held a number of leadership positions with direct responsibility for developing and launching new product lines. Prior to joining Cisco, he was a consultant at Accenture in its Electronics and High Tech Operating Unit.

Feature Your Byline

Submit an Executive ViewPoints.

Featured Event

Join the retail community as we come together for three days of strategic sessions, meaningful off-site networking events and interactive learning experiences.


Access The Media Kit


Access Our Editorial Calendar

If you are downloading this on behalf of a client, please provide the company name and website information below: