Make Room for Big Data

mobilepayThe increasing volume and detail of information captured by business, the rise of multimedia, social media and the Internet will fuel exponential growth in data for the foreseeable future.

By Barbara Brynko, Contributing Editor

Big data. If you think data is something that only big retailers such as Amazon or Netflix deal with, think again.
• International Data Center (IDC) reports that 90% of the world’s data has been created in the past two years.
• Gartner Research Group predicts that the volume of data is increasing at a rate of 59% annually.
• About 36% of retailers admit that they have customer data, but don’t know what to do with it, according to a report from Columbia University Business School’s Center on Global Brand Leadership and the New York American Marketing Association.

These three statistics represent the growing trends in big data, and it’s making retailers realize just how important big data is and how they can harvest value from their own data fields to create a better customer experience.  

“If you’re looking at the term ‘big data,’ everyone has a different definition of what it is,” said Brian Ross, president of Precima, part of LoyaltyOne analytics solution. “But essentially, you’re talking about extremely massive amounts of information coming in from many different sources.”

In order to get the most value out of such data, you have to be able to answer two questions at the get-go: “What am I trying to answer with this data, and how actionable are the analytics to get to that data.” Data can actually drive what he called the four “P”s in retail trends: perspective, products, pricing and promotions.

“Not only do we know that product needs are different depending on store location, but that a family with young children living in a suburban market has very different needs than professionals in an urban market,” Ross said. “You’re basically looking at segmenting your stores by the types of customers they serve, which usually creates customized assortments and prices that cater to the best needs of those markets and how people shop in those stores.”

If you have a loyalty program, chances are that you already have data streaming in, and if you’re not analyzing that data, you aren’t tapping the rich value of its content. Your best data may be tucked away in silos in different parts of the business operations: Structured data (customer names, addresses, phone numbers, loyalty points, purchase amounts) may be stored in one area, and unstructured or semistructured data (images, emails, tweets, Facebook Likes and product reviews) in another.

When structured and unstructured data are collected, merged, and analyzed, they can yield analytics and metrics to provide a better understanding of customer shopping habits, best-selling products, seasonal favorites, or the next big seller for your convenience store.  

“In many ways, there has never been a better time to be in retail,” Ross said. “The ability to access big data and get to a true understanding of customer needs for rich data is really enabled by a number of trends that are driving the industry.” Companies that have core transactional data and loyalty programs can attach a unique customer identifier to previously anonymous transactions. It’s not only about what’s being sold, it’s who’s buying and what their needs are.

But retailers can’t succeed with data alone. Analytics are a critical part of the equation to make sense out of the volume of incoming data. Analytics were hard to calculate years ago because it was a manual process that took painstakingly long to calculate. Now with cloud-based solutions and software, the process is much simpler. Analytics can help you understand your customers through segmentation by analyzing sales and activities within certain stores, and data visualization tools can transform huge amounts of metrics into an easily digestible three-dimensional pie chart to capture information at a glance.

Where’s the Data?
“We use data that is primarily gathered through our loyalty rewards program and what’s gathered through our point-of-sale (POS) system,” said Patrick J. Lewis, partner/CEO of Oasis Stop ’N Go Convenience Stores and KickBack Points Rewards Systems. “So most of the data we use is transactional data supplied from sales. The loyalty program supplies data too, but that has a different data set attached to it of customer information.”

Both data sets offer metrics and information that you can draw insight from and make decisions on.

Lewis is not new to big data. He’s been working with it since 1999. “We were definitely pioneers in the loyalty program in the convenience store market,” he said. What started off as a self-made loyalty program for 12 Oasis Stop ’N Go Convenience Stores in Idaho, blossomed into a sister company called KickBack Points Rewards System that now offers loyalty programs nationwide.

“There’s no small amount of irony in it either,” said Lewis. “We’re convenience store operators who know a lot about selling Twinkies and gasoline, but we couldn’t be accused of being technologists. Things certainly changed.”

Lewis said the early data that was collected consisted of basic enrollment information, or structured data (customer names, addresses, and phone numbers). He admitted that there wasn’t much fundamental data to work with, but the data they had was actionable.

Likewise, Jenny Bullard, chief information officer at Flash Foods Inc., said each one of the 170 stores in the Flash Foods network in Georgia and Florida now pushes data into its data warehouse every five minutes. “So it’s real-time information that we can access and report on,” she said. “And that data includes every item that is sold in our stores, every transaction, how that transaction was paid for—whether it was with cash, credit or debit card—and a lot of market basket information.”

Flash Foods also has a loyalty program that has been in place since 2006, which Bullard said gives the marketing department a tool to merchandise the stores and even where to place a product to make it more attractive to customers. And the convenience store chain has been using in-store scanning since 1997.
As with Oasis Stop ’N Go Convenience Stores, Flash Foods initially created its own homegrown application to gather customer data and eventually signed up for Pinnacle’s enterprise performance management (EPM) product, which translates metrics into business meaning and gives the marketing department more tools to see what the customers are buying.

“One of the biggest things that EPM brought to the table was analytical tools for our operations department for loss prevention,” said Bullard, who reports that data can pinpoint the exact pump where there was a drive-off, and transaction or item voids that may be a signal of internal theft.

Starting at Square One
At first, Lewis said Oasis Stop ’N Go’s data initiatives were simple. “We were able to find out who our top customers were, who our top 50 customers were at a particular location, how much they were spending, what their annual value was as a customer, and then create offers or promotions specifically for those top customers.”

Fast-forward to today, and the data from the loyalty program is now integrated in the company’s POS system, which provides transparency into all purchases in real time.
“That helps us create promotions and helps drive customer behavior,” Lewis said. For example, some customers may be purchasing fuel but not coming into the store to make purchases. “We can target those specific customers, give them coupons, or actionable offers to get them to come into the store.”

Details about a customer’s market basket are basic treasure troves of information, and adding surveys or databases helps convenience stores get a better picture of their customer base.

“If there’s a particular purchasing behavior I’m looking for, I can go back to the data and narrow the fields for the offers,” Lewis said. The same methodology can be used if you are trying to target customers who aren’t using certain services, such as someone who is buying fuel, but not the car wash, or eating at the attached restaurant but not purchasing items at the convenience store. “We can really dial in and find the right offers for our customers and give them more meaningful offers.”

So discounts are targeted; they aren’t offered to everyone, including those who were going to do business with the convenience store anyway.

“Now there’s one caveat,” Lewis said. “You have to have the data to begin with. It takes a little while, but the more data you have, the more time and more experience you collect, the better the decisions you can make.”

Lewis estimated that six months of transactions should generate enough data for starters, including who the customers are, their preferences, what they buy, and what they usually purchase together in their market basket. The data can provide plenty of insights, even bundling ideas.

For Bullard, sitting down with the marketing and operations department and talking to them about their needs and what data they are expecting to see can streamline efforts. “Sometimes that straightforward communication will help you hit the ground running instead of stumbling with the wrong information or showing data in the wrong format,” she said. “It can be very helpful.”

The Big Benefits of Loyalty
Loyalty programs have two key components, Lewis said. If you design a rewards program for customers, it will elicit a certain response, so they come in more often and spend money. And they’re more likely to be price-tolerant and less likely to leave you for a lower price, so you get a lift right there. But eventually, the lift hits a plateau.
Because the market is changing so fast, data has to be actionable and accessed quickly so a retailer can maintain a competitive advantage. You can launch promotions and gain insights into how to merchandise your store, where product placement would work best, and what product mix is preferred; you can even get an idea about a customers’ payroll schedule, based on their spending patterns. So timing is everything.

“The only real way to get lift-on-lift is to change customer behavior by creating promotions within the loyalty program,” Lewis said. “Look at their behavior, target certain behavior that you’re after, which we call triggering. A trigger group is the target group within a database we offer promotions to,” which can be a coupon at the bottom of a receipt or a coupon sent directly to a customer’s cell phone.

If customers use their loyalty cards at the pump, the advertising they see is specific to them. “They might not know it, but we’re handpicking the commercials they are watching when they are out there at the pump,” depending on the data, Lewis said.

Cashing in on Twitter Tuesday
Tapping social media is yet another potential avenue to grow customer data. Lewis prefers to “close the loop” by offering promotions that have a beginning, middle and an end.
Offering a fuel discount on Twitter Tuesday, for example, provides a starting point since customers have to be following on Twitter to know about the deal in the first place. “When they fuel up on Twitter Tuesday, we can match up the two and pass that data to our marketing manager for immediate feedback to adjust the offer or try something new,” Lewis said.

Stores can often see the benefits of marketing and the rewards of a loyalty program, but the power to help retailers calculate individual customer-level price optimization is often overlooked. It’s what Lewis calls “pricing elasticity.” If every customer who walks through the door is getting the same price, it’s inefficient, he said. The price is too high for some customers and too low for others. “If you are pricing too low, you are really leaving money on the table.” A loyalty program lets retailers offer value where it’s needed.
The key is keeping customers happy. “Convenience stores aren’t known for their marketing strategies,” Lewis said. “Typically, you build a store, and it’s all about location. Customers are coming to you.” But even at the most basic level, data from a loyalty program can help a store provide a good return on investment and keep customers happy. “If you don’t use the data to build customer-specific promotions, you are missing out on opportunities.”

A Little Privacy, Please
Most retailers agree that acquiring customer data comes with provisos, which is perfectly understandable.

“We have an incredibly strict privacy policy with our customers,” said Lewis, where customers must opt-in and all of their data is secure. “We want to
provide good value for customers without looking creepy.”

For example, suppose you’re a Diet Coke drinker fueling up at Pump No. 5 when a Diet Coke commercial appears on the screen. Are you connecting the dots that your private data is being used to entice you, “or are you thinking, ‘Ah, 50 cents off a Diet Coke and I’m really thirsty,’  It’s all a matter of walking that fine line between providing a service and invading a customer’s personal space,” Lewis said.    

Respect that your greatest asset is your customer and that his trust is paramount. Most customers are OK with a retailer using their data as long the retailer provides a relevant user experience. “You have to be clear about all the customer data privacy and how you’re going to use it, and you always have to abide within those rules without exception,” Ross said.

The ABCs of Big Data
For retailers who want to jump on the big data bandwagon, Lewis suggested going through the due-diligence process and checking out best practices. The Tech Event is also an option, where prospective clients can check out several companies that offer services, consulting and expertise in one place. He considers these types of technology conferences a great spot to chat with representatives from different companies in order to get a high-level understanding of what each offers.

With retail moving at such an exponential pace and the rules of the game changing, “you can’t be doing business out of a cigar box anymore,” Lewis said. Everyone is going to be using data to make better business decisions, and loyalty programs can help gather that data and deliver offers to customers.
“Competitively, you’re at such a disadvantage if you’re not getting that data, finding out who your customers are, and marketing to them more intelligently.”
Bullard admitted that the granularity of the data and the information it supplies is a big help in creating marketing strategies. “I constantly have our director of marketing asking me how we ever survived without the information we have today,” she said. “We have a lot of information to drill down into that gives us some best practices and understanding of how we want to look at the data.”

And Ross agreed: “Companies that start making the most of their data are going to win, and those that don’t are going to have a hard time. The reality is, it’s not a question of ‘if’ you’re going to start using your data wisely; it’s a question of ‘when.’”



  1. Barbara nice article on Big Data. Designed by data scientists, HPCC Systems is an open source data-intensive supercomputing platform to process and solve Big Data analytical problems. It is a mature platform and provides for a data delivery engine together with a data transformation and linking system. The real-time delivery of data queries of the Roxie component is a big advantage for marketers needing to take action from data insights. More info at

Speak Your Mind