It’s hard to read a business publication these days and not come across an article about big data. It’s the hot buzzword du jour, and depending upon the source, can be regarded as the “next big thing,” or an overhyped gimmick, or a company’s key to having an edge over its competitors.
The sheer amount of data now available is mind-boggling. According to The Economist, Wal-Mart conducts more than a million customer transactions every hour, and its databases consume more than 2.5 petabytes of information. A petabyte equals a quadrillion bytes of information, or as The Economist points out, 167 times the number of books in the Library of Congress.
The problem is, there’s now so much data available it’s becoming inaccessible. How can businesses make sense of all this information?
In an effort to sort out how big data can help businesses, Texas CEO Magazine brought three leaders of Texas companies together for an Enlightened Speakers’ Series presentation in Dallas. Bob Zurek, the Senior Vice President of products at Epsilon, a Dallas-based marketing services firm and division of Alliance Data; Alex Siskos, Vice President of Crossmark, which tracks sales data for over 100,000 consumer products, store by store and item by item; and Becky Gomez, marketing communications manager of Half Price Books, each provided a different take on how their companies are using big data.
Zurek had an example of the way Epsilon uses its data.
“We have an email platform that’s responsible for delivering over 14 billion permission based emails annually,” he said. “I can walk into the data scientists and I can ask, ‘We have a client that wants to ship an e-mail to males, aged 18-21, within a 25 mile radius of Boston. What time should we ship those emails?’” Epsilon tracked the click open rate for the emails, and found the best time for that group is between 8:31 a.m. and 8:45 a.m. “The client found a difference of 80 percent more open rates on their emails because the data scientists dug deep to find that information,” Zurek said.
Siskos used an example from 7-11 stores.
“They created analytics to look at the sales of their blueberry muffins based against the temperature,” he explained. “What they realized was, when the temperature dropped below 52 degrees, the sales of blueberry muffins go through the roof. I don’t know why, but that was an interesting insight.” He said 7-11 created a production line in markets like Chicago that had cooler weather, and figured out their supply chain to make more blueberry muffins when the temperature was going to be below 52 degrees.
“Data comes from everyplace,” said Siskos, “and what I do is organize it.” He said data about consumers comes from stores, while data about retailers comes from companies like RetailLink for WalMart, Shiloh, Atlas, and Market6. “All these companies are in the business of delivering this information and then we have a conversation about how to incorporate it in your business,” he said.
Siskos said a company entering the field of big data needs to understand three things about data: “Identify, understand and act.”
“At the ‘identify’ mark – the top line view of the business – take a poll and understand what’s going on,” he said. “At ‘understand,’ we get into early diagnostics. Diagnostics is like triage – we prep the area for the doctor to come in by taking vitals and getting to the core truths about your business. Once the doctor comes in, we do category management insights and shopper insights – that’s where the activation comes in.”
Siskos said the most important piece of information he could give is to look at ways to connect data back to the process of doing business. It must fuse it into the way a company operates every day. If not, work with data will fall apart.
Siskos explained that the most popular way to tie data to a business is through a “volume decomposition report.”
“It takes your total dollars and shows you where you’re hurting,” he said. “It’s a very quick visual way to see where you have to go next.” He said Crossmark issues the report on a weekly, monthly and quarterly basis. “What we look at is no longer data – but the core truths,” he said. “Five years ago looking at this was a luxury if you could do it; today, looking at these core truths are the table stakes.”
Siskos said a company that can’t do this, can’t do the next step: diving deep into a category or retailer and telling them why they missed the numbers, what’s the right price point, what’s the right promotion and finally, what does the shopper think?
“We put all of that together through analysis and see even more – now I organize into clusters and segments to show opportunities,” Siskos said. “We create a scorecard, track the results across the country and then we start all over again with a new set of facts.”
For Gomez, big data has great potential, but Half Price Books decided to move into it slowly. The company has a 3.6 million-book inventory, and for a long time, did not have a searchable database.
“We do now,” she said. “We have tried to take it slow in order to process the data we have. Our IT department has a strong set of programmers who have built our inventory system by hand internally.” HPB’s strategy is to deal with inventory coming in by the minute. “We need to capture that and understand it the most,” Gomez said.
Businesses can then start looking at the next step in big data, added Siskos. Looking at a company’s own business needs, he said, is a myopic approach. “Instead of just caring about my product,” he said, “I started thinking about the aisle, the shelf, what products are next to me on the shelf, how those items interact with the person who’s shopping, what’s the shopper’s purpose, then thinking about how this product plays on this aisle and how does it all fit together?”
But then, companies must consider the online factor, and get beyond worrying about retail space and competitors on the same store shelf. For example, he said, 33 percent of diapers are no longer purchased in a brick and mortar store because they’re being bought on the web. The newest wrinkle is called “click and drive.” That’s when customers order online but pick up at the store. “Another version of ‘click and drive’ is to go to a locker at a central location, put in a code and pick up my groceries without ever going into the store,” Siskos added.
Feeding these new ways of retailing are the massive amounts of data that are collected continually. Siskos’ company, Crossmark, tracks 100,000 stores in the food, drug, and mass merchandise fields. “We track over 95,000 digital SKUs and add on top of that 52 weeks of data,” he said. “Then we drill down to stores, then to daily sales and sometimes drill down to data by the hour. It moves very fast.”
The analysts who handle this data are an elite force. Gartner predicts 1.4 million jobs in big data by 2014, and Zurek said there aren’t enough trained professionals to fill that gap. “The compensation in this area is going to skyrocket,” he said. “The data scientist is now the hip, new job and people with rock solid experience are going to be costly.” Zurek urged companies to move quickly to get a team of data analysts together to examine how to apply big data inside their businesses.
Gomez said Half Price Books has done that in a way that fits well with the company’s philosophy. “Nobody else does it the way we do, so when I speak with our programmers, they ask me what is it you need?” she said. “Will you use these reports week to week and month to month?” Gomez said HPB looks at transactional data from a promotional standpoint.
“We use email and social media outreach,” she said. “We are able to look at the best times to send emails, the best subject lines, reengagement campaigns for those who didn’t respond to a previous promotion. When someone signs up we have immediate responses and that helps us keep the customer engaged and they know we’re listening to them.”
Gomez said Half Price Books is very careful to pick the social media strategies that work for its business. “We grow our email list by 14,000 per month and our ‘likes’ on Pinterest, Instagram and Facebook are growing exponentially,” she said. “All of this is generating traffic to our website, as well.”
Compared to other businesses, Gomez said HPB operates on a shoestring budget. “While all the data is coming at you, keep your company in mind and be realistic about what you can take on,” she concluded.
Siskos recommended prudence as well, although on a different scale. “The data keeps growing and we’re hungry for more,” he said. “The issue we have is the 30 to 40 phone calls a week I get telling me the next ‘cool’ thing is out there, so in addition to doing the analysis, I have to figure out what companies should be looking at next.” The challenge, he said, is to be interactive with the customer, to sit down and in a matter of five minutes, drill down through the data to show where the opportunities are and what to do about them.
A member of the audience asked about security measures to protect all this data, and the panel agreed it’s a major concern. “I’m very paranoid about data security,” said Zurek. “We see all kinds of intrusions into our business because we house some of the largest amounts of consumer data, so we’re attacked all the time.” The FBI, Secret Service, and his company’s own internal security officers are always busy fending off attacks. He said the struggle with database technology is that it might not be secure enough.
Another question from the audience summed up the dilemma over big data: If pollsters can survey 1,000 people and tell us who the next president will be, why does it take billions of data points to drive your business? Does it really give us more good information or more crap? the questioner asked.
“First,” said Alex Siskos, “there’s a stat that says 80 percent of statistics are made up. Right now there are over 120,000 people in stores right now putting things into their proper place waiting for someone to come by and buy it. I have to rely on facts that come from activity showing there is a sales lift on Tuesday and Wednesday that will give you a return on your investment – that’s not statistics, that’s facts.”
Thanks to our Speaker Series Sponsors:
#TexasCEO's 10 Most Read Articles Of 2017 #2 The Changing Of The Guard: San Antonio’s #Economic #Development Future texasceomagazine.com/features… @SanAntonioEDF #1 Deep Roots In The Heart Of #Texas: The Kaspar Family’s #Holistic Approach To #Ranching texasceomagazine.com/features…
#TexasCEO's 10 Most Read Articles Of 2017 #4 @Jim_nyquist: At The Inflection Point Of #IIoT texasceomagazine.com/features… #3 Building #Innovation: @turner_talk Builds On Empowering Individuals texasceomagazine.com/features… @Turner_DAL @TurnerSouthTX #MiddleMarket #entrepreneurs
#TexasCEO's 10 Most Read Articles Of 2017 #6 The Cilantro Diaries: #Business Lessons From The Most Unlikely Places texasceomagazine.com/book-rev… @lgomez123 #5 From #Franchisee To #Franchisor: How Gordon Logan Built @SportClips texasceomagazine.com/features… #MiddleMarket #entrepreneurs
#TexasCEO's 10 Most Read Articles Of 2017 #8 #Mexico Is The New #China by @DrPippaM texasceomagazine.com/departme… #7 Growing The Next Generation Of Texas #CEOs: #CEO-to-CEO #Mentoring texasceomagazine.com/features… #Texas #MiddleMarket #business #entrepreneurs