Boosting Store- Level Performance through Big Data
“Big data can help every location understand its unique strengths and weaknesses, and then course correct to improve financial performance metrics”
Many retailers use CRM-related big data to segment their customers into clearly defined groups. This provides opportunities for precision marketing to customers based on their current value to the company, buying patterns, preferences and even loyalty card use. Typically, this data focuses on the customers themselves and is especially pertinent to marketing teams. However, corporations also have another vital data asset: Information about the performance of their physical locations.
Data collected about the actual location—ranging from operational performance to site characteristics to customer satisfaction—can be used to increase storelevel performance on any number of dimensions. Retail brands need to know that their stores are executing on the brand promise and delighting customers. Big data can help every location understand its unique strengths and weaknesses, and then course correct to improve financial performance metrics, such as year-over-year growth in same-store sales, basket size and conversion rates. Let’s take a look at understanding conversion rates—an absolutely critical metric.
Start by Taking Stock of Conversion Rates
Thousands of customers walk through the doors of retail stores on a daily basis. Retailers use their CRM databases to drive that traffic with precision marketing campaigns. Those campaigns include everything from new merchandise to promotions, and often mean six to seven figure investments from the marketing budget. But, once a customer walks through the door, only some of them actually make a purchase. Retailers’ ability to increase the number of customers who purchase—the conversion rate—will ultimately make or break their success. Big data can help retailers identify exactly the right levers to increase conversion rates at every location. Imagine the impact of increasing conversion rates from 20 percent to 30 percent, or 40 to 50 percent. It literally equates to millions of dollars in sales.
To understand conversion rates, retailers should consider four different data streams: operational excellence (obtained from audits, mystery shopping, price checks, etc.), the specific characteristics of the location (square footage, number of checkout stands, urbanicity, etc.), and two data streams that we’ll explore in detail: customer experience feedback (structured and unstructured) and measures of traffic/ footfall.
Virtually every retailer measures customer feedback using structured surveys. Customers are invited to participate in these surveys via email push campaigns or by including a survey invitation on a POS receipt. However, these surveys have a fatal flaw. Only those customers who made a purchase provide feedback, as they are the only ones who can be identified. The experiences of those customers who browsed and did not purchase are rarely captured— and they are the key to improving conversion rates. How can retailers overcome this fatal flaw?
First, invest in systems that accurately count store traffic. These systems capture the number of customers who enter and exit a given location using door sensors and video. Next, match the number of customers to the number of purchases. If there is an average of one purchase for every five people who enter the store, the conversion rate for that location is roughly 20 percent. This will become your conversion rate metric, and that metric will vary from location to location. Some may have conversion rates at 20 percent, some at 50 percent. Conversion rates can vary wildly for multi-location retail businesses. One hundred stores can have rates hovering around 60 percent, while 400 stores have rates closer to 20 percent. The goal will be to help every location maximize its conversion rate by improving operational excellence and customer loyalty
Surveying the Customer (and Non-Customer) Experience
The next step is to supplement the surveys given to purchasing customers with data from customers who did not purchase. Both points of view are required. To obtain information from customers who only browsed, consider using QR codes within the store, self-serve kiosks, mobile applications that deliver surveys based on location within the store, and customer intercepts when customers exit the store. All of these methodologies can capture the experience of customers who do not actually purchase and will be vital to understanding conversion rates. When it comes to designing the surveys, the questions should seek information that the brand needs to understand (merchandise selection, value, etc.), as well as items that each store controls within its four walls, such as knowledgeable and friendly sales staff and whether the checkout process was hassle-free.
Modeling and Measuring
Now big data can come into play. Predictive conversion rate models using the wealth of location-specific information can be created using multiple data streams that measure operational excellence, customer experience, social media and site characteristics. These models will identify the levers that locations can pull to improve their performance. The levers could include using the planogram correctly. Or pointing out that a wireless store needs to improve adherence to training guidelines for conducting phone demos and explaining rate plans. At a home improvement store, a lever might be increasing the checkout speed by having more lanes open.
Big Data for Action and Accountability
Big data is only valuable when it is put into practice. Every location will have opportunities to improve their performance on any one of the levers that improve conversion rates. The store manager needs to own the plan for improvement with clear actions to solve for gap to goal on both the levers that move conversion rates and the conversion rate itself. Holding locations accountable for change is the most difficult aspect of any program focused on increasing conversion rates. The analytics team can crunch numbers and create models all day long, but unless store managers act on the insights, conversion rates will remain flat. In an industry where margins are fine and competition is intense, the sub-par performance of a handful of stores can be damaging for a brand. Big data makes it possible for multi-location retailers to make fast and informed decisions that optimize every location’s ability to increase conversion rates, cut out unnecessary costs and deliver on the brand promise—the underpinnings of long term success.