Return fraud and abuse continues to cost retailers billions of dollars every year—$16 billion to be exact. But some retailers like Finish Line, a $16 billion premium retailer of athletic shoes, apparel, and accessories, are fighting back and combatting return fraud head on with winning results.
Headquartered in Indianapolis, Finish Line operates approximately 1,400 branded locations primarily in U.S. malls and shops inside Macy’s department stores and has a strong online retail presence. With a mission to connect to young, fashion-conscious individuals through a premium-brand experience, Finish Line’s retail stores play an important role in delivering this premium experience. More than 14,000 “sneakerologists” are employed across the country, so maintaining a consistent loss prevention strategy and process is paramount to preserving that premium customer experience.
Seeking to protect the company from return fraud while continuing to provide an inviting customer experience, in 2005 Finish Line turned to a real-time, consumer-based return authorization solution called Verify from The Retail Equation. It helps prevent fraudulent and abusive transactions from occurring by using predictive analytics to identify and stop high-risk returners, increasing net sales, and reducing return rates, while ensuring all legitimate returns can be made.
Creating an effective returns policy that protects the company from loss without negatively impacting sales or customer loyalty requires a careful balance; it’s both an art and a science. This solution quickly became a valuable tool to streamline the return processes at the point of sale, help enforce the return policy, and provide management discretion to evaluate suspect returns as needed.
A Powerful Resource for the Whole Team
Understanding that the return authorization solution could bring far more benefit to the company than just within the loss prevention department, we collaborated with Finish Line’s IT and operations departments through the selection and implementation processes. Our LP team had the foresight to know that a technology spending decision should not be made solely through the scope of loss prevention because the success of this implementation would depend on—and benefit—virtually every department.
Several years after we implemented the returns management technology, our foresight has proved to be true. By understanding return patterns through the data captured in the system, we have learned about merchandise categories, product groups, and specific items that can help loss prevention, merchandising, operations, and even our vendors. We have a tool that can truly curb return fraud without alienating loyal customers and deterring future sales. And we were able to integrate the system with our existing exception-based reporting application to provide a clear picture of what’s happening in the business.
Furthermore, for operations executives and in-store teams, the solution has helped tighten POS processes and controls. In effect, we have helped improve the consumer’s return experience and engender their loyalty through a quick, efficient, and easy-to-manage return process that eliminates all guess work from our store employees.
A Historically Strong Offense against Return Fraud
The solution helps quickly determine whether a return is accepted or declined—guiding the salesperson and removing the guesswork. While each retailer can choose the best way to identify consumers, previously with every return at Finish Line, the customer was required to present a valid government-issued identification, which was then swiped at the point of sale (POS). Pertinent data was captured in the system, encrypted for security, and instantly run through a series of predictive algorithms and statistical models. Almost immediately, the system generated one of three recommendations for the return—accept, warn, or decline.
The automated returns tool improved the consistency in Finish Line’s return procedures. The solution’s analytics backbone consistently uses the same parameters to apply the return policy guidelines, as opposed to a cashier having to make a subjective decision with each return—a situation that can escalate emotions and lead to customer dissatisfaction or accepting fraudulent returns in a panic, both of which are very damaging to a retail company.
Taking a proactive, analytics-driven approach to return authorization helped deter fraudsters and shoplifters who often attempt to return stolen merchandise for cash. Any attempt to return merchandise with frequency is detected, and immediately the transaction is flagged, and the store associate is notified to deny the transaction. Prior to this system, we were unable to capture that data, much less make it available in real-time.
With this increased insight and intelligent decision making come cost savings. We recognized a return on investment through an improved rate of return within one year of implementation, and we expect to again see a decrease in return rate year-over-year when we re-deploy the solution with our new POS system. Shrink was also down markedly over the same time period. Additionally, there were improvements in customer experience, greater POS efficiency, and greater profitability thanks to reduced loss.
Leveraging the Future to Create More Loyal Fans
The most exciting part of Finish Line’s new solution deployment is our ability to directly enhance the return experience for our loyalty customers. Very soon, members in our Winners Circle program will need to present a government ID at return only once, and then every time afterward they will be identified by their loyalty info. This provides multiple benefits:
- It makes the return process easier and more streamlined for Winners Circle members,
- It creates an “easier returns” selling benefit to attract and sign up new loyalty members,
- It gives Finish Line associates an additional reason to ask a customer for their Winners Circle number at time of purchase, which gives us even better overall purchase history,and
- The collection provides more big-data transaction information tied to specific consumers to help my business partners in store operations, marketing, e-commerce, and merchandising.
By capturing this data in real-time at POS, we will provide greater insight into shopping trends and patterns and even better strategies to best reach customers and maintain loyalty. Additionally, an intelligent return authorization solution can help treat different consumer segments uniquely. We could add even more value to our loyalty program when these best shoppers can be quickly recognized and given a more lenient return policy. Ultimately, it creates a more premium return experience for Winners Circle customers.
We are also creating an omnichannel environment by building a seamless return process between all business channels, as well as an expedited return/exchange process at the POS for both the customer and associate. Imagine implementing a “buy online, return in-store” policy with an easy way to validate that the purchase was actually made from your e-commerce site. This is very important for Finish Line since a majority of returns made from e-commerce sales are performed at stores and not sent back to the distribution center. The risk of these returns is greatly diminished, and the store can now focus on the shopper experience as opposed to fraud mitigation.
These features are just scratching the surface. Knowing shoppers and their behavior in real-time, in-store is a huge advantage that adds value to the business.
The Power of Connected Big Data
When it comes to gaining insight into shopping habits and trends, there is no question that knowledge is power. Finish Line and other retailers are embracing data tools like exception-based reporting, predictive modeling, digital video, and ROI analysis to help identify previously hidden relationships to take decisive action to improve customer service and reduce the specters of fraud and shrink.
Here’s how it works. Using analytics on top of linked data, we can build connections that reveal more about the customer. The better informed a retailer is about its shoppers, the better it can create strategies to market to them to generate future sales that meet the buyers’ needs and expectations, and therefore minimize returns. In Finish Line’s case, detailed data captured during a return transaction can ultimately reveal trends and preference that can help us provide better incentives, greater customization in the shopping experience, or additional merchandising flexibility.
For example, let’s say a new shoe is hitting the market that is flashier and more colorful than shoes in the past. We decide to merchandise the shoe prominently on a display at the front of the store. As a result, the new shoe flies off the shelves, but then gets a lot of returns. The return data—what sizes, colors, styles were returned more—can be analyzed to separate potential fraud from other consumer issues, and ultimately provide insight to other departments to help determine, for instance, whether certain colors are less popular and should no longer be carried, or how merchandising displays might be designed differently to entice and help the shopper make a lasting purchase.
Our primary objective with implementing this technology was to create a better consumer experience in our stores, while at the same time to get a handle on fraudulent returns, which are expensive for retailers. We then started to see other opportunities emerge by looking at the data captured in the system to help modify other processes by adjusting to meet customers’ interests and expectations.
Collecting data and impacting business on a greater scale creates significant benefit. At Finish Line, the visibility all starts with the customer’s return. Over time you learn enough to link all of the data and have a holistic view of the customer. It is then possible to drill down even further through common threads of data to link disparate points.
Each data type has variables allowing it to be linked with other data types. As shown in the customer linking details chart on page 19, beneath each high-level data type listed as a column heading are keys that you can link from one type of data to another. For example, employee number can be easily linked to customer name and then to other disparate data. The ability to take this additional information and tie it into existing and future analytics tools is very important.
For Finish Line, these links add significant value to the analysis effort already in place with our exception-based reporting tool. In one case we started with a SKU that was missing in a physical inventory. We linked that SKU to a refund transaction during that inventory period and then connected it to the employee performing that refund.
With infinite ways to use this data to improve business and gain intelligence, we have a tremendous cross knowledge of returns with inventory and employee behavior. To fully realize the benefits, it’s important to educate your business partners in store operations, marketing, e-commerce, and merchandising about customer behavior. With everyone on the same page, the real benefits are achieved, including an improved customer experience while shaping customer behavior.
The Benefits to the Bottom Line
The more you know about a customer, the more you can affect your bottom line. Authorizing returns may have started as both a customer service improvement and fraud prevention initiative, but it has proven to be so much more.
Everyone in the organization will benefit from this scientific, analytics-based understanding of overall customer behavior. It’s important to increase awareness about our overall return activity now and as our policies and procedures change. Store operations can reinforce policies with all store associates and be actively engaged in the shaping of new, profitable customer behaviors. Marketing is enhanced with a better loyalty program. Merchandising can be more creative. E-commerce is streamlined.
At Finish Line this is much more than a loss prevention initiative; it involves the entire company. And the entire company has and will continue to benefit greatly from big data.
SIDEBAR: Big Data Is Reshaping Retail LP Operations
by David Speights, PhD
Businesses, including retailers, are losing 5 percent of revenues to fraud every year according to the Association of Certified Fraud Examiners’ 2014 Global Fraud Study. This is a staggering number, especially when applied to the 2013 estimated gross world product, which translates to nearly $3.7 trillion in potential projected global fraud loss.
Many retailers are taking action to combat fraud by employing new technologies to monitor high-risk retail transactions. This is helping reshape retail loss prevention operations to deliver a better customer shopping experience, while effectively protecting company bottom lines.
Technology Enhances the Collection and Monitoring of Data
Retailers collect data from many sources, including store sales transactions, store video, traffic counters, alarms, merchandise movement, loyalty programs, e-commerce click paths, and much more. A large retailer collects millions of transactions and hundreds of millions of line items per day. To that, they add 30 to 60 gigabytes of video per store, per day. For a 1,000-store retailer, this could total 22 petabytes per year, or the equivalent of 23,068,672 gigabytes.
Conventional systems like exception-based reporting and data-mining systems uncover direct relationships that occurred in the past on a single identifier. But big-data analytical tools take analysis to a new level by detecting the connections among seemingly unrelated identifiers to reveal underlying larger groups of transactions and individuals. For example, a return authorization technology like The Retail Equation’s Verify-3 solution helps prevent fraud and abuse in real-time during the return process, not the day after, while simultaneously linking in all related information to an individual. This type of response is not possible with conventional systems because they simply cannot process the complex analytics and deliver accurate answers fast enough to authorize a transaction in process.
Many companies and solution providers have approached the data size or analytics problems by investing in a small number of bigger, faster hardware machines. This works up to a point. However, the massive amounts of data building up in the system and the complex analytical methods required to unearth the information is more than most traditional computing architectures can handle.
Big data-oriented companies achieve high-processing speeds by using special tools like the Hadoop platform to split data into thousands of chunks and distribute the load across a very large number of machines. We have developed custom software to operate in this environment to tackle very tough analytical challenges. Additionally, tools like IBM’s PureData (formerly Netezza) data-warehouse appliances are also used for added throughput and operate on a similar parallel-processing architecture. This architecture decreases processing time by more than 90 percent versus more traditional hardware/software deployments. A query that may take five hours to process using SAS on a conventional single-server system takes only minutes on a Hadoop or PureData parallel-processing architecture. In the field a return authorization completes in milliseconds.
Knowing Precisely What to Look For
Predictive algorithms and machine learning techniques rely on big-data tools to quickly improve the shopping experience and reduce return fraud and shrink simultaneously. Companies can process the data from all the transactions in the chain and identify suspicious behavior indicative of any form of return fraud or abuse, including wardrobing, returning stolen merchandise, receipt fraud, price arbitrage, price switching, double dipping, check fraud, and tender fraud. When an individual attempts to make a return, systems perform calculations and in a fraction of a second predict the likelihood of whether the return is fraudulent.
While the vast majority of consumers (about 99 percent) are approved, those whose actions are highly suspicious are warned or denied. This is an important point to notice. The system allows and supports generous return policies, so profitable consumers enjoy a fast and pleasant return process, including those who make numerous returns. In fact, the most valuable consumers tend to have a very high number of returns, which is why it is best to not rely solely on simple return-velocity calculations, but rather use big data to identify fraudulent patterns of behavior in real-time.
Complex queries can also be used to identify organized retail crime rings and fraudulent returners by linking seemingly independent events. The diagram above shows a cluster of suspicious purchase and return events.
At the center of each dandelion-like cluster is one person (or one ID). The thin lines connecting the clusters show a “hidden” connection, such as a gift card being passed among conspirators. On their own the clusters may appear to be legitimate (high-volume buyers often return many items), but the value of the returns exceed the value of purchases, and the connections indicate probable fraud by a group or an individual using multiple identities. Software-as-a-Service applications can halt the group’s returns immediately, and special reports can provide investigators with the information necessary to pursue a case. The big-data tools allow companies to maintain and update the linked identities on more than a billion linkages each day.
The amount of return-related fraud is a staggering $9 to $16 billion problem, according to the 2013 Consumer Returns in the Retail Industry report released in January 2014 by the National Retail Federation. Many of these losses are preventable using the technology available today. Fraudsters depend on system delays and lapses in judgment by the cashiers and associates on the front lines.
However, when big-data analytics replace subjective decisions, fraud and shrink diminish substantially, reducing return rates by an average of 8.2 percent and shrink by 12.95 percent. A $2 billion retailer would see about $15 million in savings per year, and retailers see a steep decline in return rates beginning immediately after the system is live.
