Read an Excerpt



Read Chapter 1 of Larry Freed’s new book Innovating Analytics below, or click here for a PDF download of the Introduction and Chapter 1.

Chapter 1


In the fall of 2011, ForeSee played host to a few hundred clients who came to our hometown of Ann Arbor, Michigan, for our annual user summit. The week of our user summit is always such a great time to interact with our clients and hear their opinions, not only about what we’re doing but also about the customer experience industry in general. Among many other topics up for discussion over the three days, I was planning to introduce WoMI, or the Word of Mouth Index, which my company, ForeSee, designed to substantially build on the value of the Net Promoter Score (NPS). At that point, we’d already conducted research to test the fundamental concepts behind WoMI and were ready to invite our clients to join in for the second round of testing.

I didn’t have long to wait to start hearing attendees’ opinions. In the lobby, on my way to grab an early breakfast in the University of Michigan’s Executive Education Center, I passed one of our clients on his way back from the exercise room. (I wish I could say I ran into him in the hotel gym, but I can barely make time to eat during our summit, much less exercise!)

As we crossed paths, he caught my eye and bellowed, “Hey, I hear you’re gonna tell us tomorrow all about why you hate NPS! Let me tell you something. You’re way off base. I love it.”

I smiled, not quite prepared for a confrontation at 7:24 in the morning, and replied, “Well, you’ve got me wrong. I don’t hate NPS, although in the past I’ve had critical things to say about it. But I’ve come to recognize some of its strengths. Why do you like it so much?”

John was a CEO at a multichannel retailer. It was his first year at our summit, though his staff had attended every year the company had been a client. As the company struggled to make sense of how various customer touch points interacted with each other and impacted the overall customer experience with its brand, John had come to sit in on some of the higher-level strategic sessions at our summit. I’d been told he was looking forward to hearing our take on NPS, which he enthusiastically endorsed in investor and analyst calls as a critical metric for his company. John explained that he loved the simplicity of one question.

“All you need to do is ask, ’Would you recommend us to a friend or colleague?’” he said. “Then you just find out your score and whether your customers are detractors, passives, or promoters. Of course you know all this. But what I really like is that it helped me line up my entire organization to focus on the customer experience. I know you say the margins of error are high, but who cares? It’s a single number, and it’s directionally accurate. It doesn’t need to be exact.” He shrugged. “So how can you be so against it?”

Normally, I’m the first guy in line for a rousing debate about NPS, but I had exactly six minutes to get a bagel before I had to be at my first pre-summit morning meeting. I sidestepped an answer by politely laughing and saying that if he would attend my presentation the next day, he would hear my concerns about NPS. And he would hear how WoMI was a next-generation approach for companies who wanted to take their measurement of word of mouth to another level. John agreed to listen, and I agreed to further discuss my ideas with him after my talk.

My next opportunity to meet attendees and listen to their concerns came later that same morning between sessions during a snack break in the atrium. I ran into Anna, a vice president of customer experience at a packaged-goods company who was not as glowing as John was about NPS. Anna admired certain features of NPS, particularly its seeming simplicity, but was having trouble making it actionable in her company.

“It’s great to have this one number,” Anna said, “but what do I do about it? So I know that some of my customers are detractors, but how do I turn them into promoters? And are detractors really out there bad-mouthing us, or are they just the kind of people who never recommend anything?

I nodded in agreement since I had heard these concerns many times before and, in fact, had substantial research that I intended to present the next day that confirmed Anna’s suspicions that you can’t accurately measure or predict detractor behavior by asking only how likely someone is to recommend something. Like John, I told Anna I would be glad to discuss her situation with her after my talk.

As I listened to the speakers on the first day, I was reminded that both John’s and Anna’s opinions had to be viewed in the context of how far and fast the fields of customer satisfaction and customer experience had come in only a few years. Providing a good customer experience had transformed from an ancient, always acknowledged but soft goal of every company. The growth in the use of websites and mobile apps allowed companies to track just about every action a consumer would take and where analytics were making it scientifically measurable. In my mind, the major competitive advantage that companies have in an era where innovation in products and services is increasingly difficult to achieve is providing a superior customer experience. As I shared in my first book, Managing Forward , when you collect and calculate customer experience data the right way, it is possible to predict with a surprising degree of accuracy a company’s future success or failure.

The data revolution in business is evident in many other fields. Take baseball and politics.

If you have read Michael Lewis’s book Moneyball or seen the movie based on it, you are already familiar with the impact of the new stats (or Sabermetrics) on America’s pastime. From the game’s earliest days, the numbers that counted in assessing a hitter were batting average, a simple compilation of the number of hits per times at bat; home runs, a total; and runs batted in, where anything over 100 was considered Hall of Fame level. A .300 hitter, three hits every 10 times up, was the rarest of talents and eagerly sought after by every major league team.

That all changed when Billy Beane became the general manager of the Oakland Athletics. With a very limited budget by baseball terms and forced to play against teams such as the New York Yankees and the Boston Red Sox, whose payrolls could approach and exceed $200 million, Beane had been drawn to the analytical work first done by a baseball fan named Bill James in the late 1970s and 1980s for new ways to gain a competitive advantage. Within a few years, James’s groundbreaking analysis had begun to claim believers among a few baseball executives.

One of James’s major arguments was that on-base percentage, the amount of times a player reaches base by either a walk or a hit, was actually a much greater factor for predicting team success than batting average. Put simply, the more runners who make it to base, the greater the likelihood a team will score runs—and runs win games. Beane adopted the Sabermetrics approach and has constantly produced winning teams and play-off contenders for many years. The analytics developed by James have now been refined by others in baseball and have become increasingly sophisticated. Analytics have spread to other sports, as any follower of ESPN like me can attest, and to the arguments fans use as they voice their opinions on sports talk shows. Even MIT now hosts a sports analytics conference every year.

A similar revolution has occurred in politics. In the 2012 presidential election, the Obama campaign focused on analytics to drive decisions about messaging and marketing. Each night in the final stretch of the race, Obama’s analytics team ran 66,000 simulations through its computers to have a fresh perspective on the battleground states. The real-time data then drove decisions on how to spend money and make it count. “We were going to demand data on everything; we were going to measure everything,” Jim Messina, Obama’s campaign manager, said. Whether optimizing e-mails, building polling models, developing a communications strategy, or creating a social media army, analytics gave the Obama campaign an edge over that of its competitor, Mitt Romney, in an incredibly tight race for the popular vote.

The predictive power of analytics in politics was shown in even sharper relief by the prognostications of the New York Times’s Nate Silver. A big fan of Sabermetrics, Silver correctly predicted the election results of every single state, including the overall Electoral College totals. It was a dazzling display of the new analytics, trumping the observations of former Republican presidential advisor Karl Rove and the more traditional analytics of the Gallup organization, both of which predicted the election for Romney.

Analytics are a moving target; where you sit in time determines your sense of their power and utility. In the 1950s, New York Yankees manager Casey Stengel found that RBIs were a meaningful and useful stat; in the 1990s and 2000s, for Yankee skipper Joe Torre—not so much. When Boss Tweed was counting votes at the beginning of the twentieth century, all he needed was a pad and a pencil; for David Axelrod, Obama’s chief advisor in the past two elections, his BlackBerry was capable of absorbing all the data flying at him.

So it was not surprising that John and Anna had different takes on NPS: John, the head of a company, looked for ways to simplify an already overcrowded score sheet of Key Performance Indicators (KPIs); Anna, a vice president, tried to find approaches to implement a concept she didn’t quite have total faith in while still moving the needle on actually improving the customer experience. Thus it came as no surprise to me when that evening, at my favorite event of the summit (a huge party in the University of Michigan football stadium, aka the Big House, that has become an annual tradition), I was buttonholed by an analytics manager who had the exact feelings toward NPS that John had assumed I had—he really hated it.

Alex led a team of analysts at a Fortune 500 financial services giant who managed all the behavioral and attitudinal data coming in about the company’s web and mobile channels. Alex was fuming about something I had long been aware of—that the whole idea of “the ultimate question” didn’t make much sense.

“Just for starters,” said Alex as we waited to order our first beers (Bell’s Oberon of Michigan, of course), “NPS doesn’t distinguish between positive and negative word of mouth, nor does it differentiate between passive and active word of mouth.”

Downing a handful of peanuts, Alex launched into a full frontal attack on the NPS methodology, including the dangers of reducing a 10-point scale to a 3-point scale, which he contended greatly increased the margin of error and eliminated subtle but important differences in customer behavior.

I told Alex that I would be addressing his concerns the next day and giving my take on both the advantages and the disadvantages of NPS. I also told him that I would be introducing both a new metric that greatly enhanced NPS called WoMI, which would go a long way toward alleviating his concerns, and a more complex model of the Customer Experience Measurement Ecosystem.

Accelerated Darwinism   

Despite my attempt to delay responding to Alex for a few hours, as I drank my Bell’s Oberon I couldn’t help expressing some of my next day’s comments. I agreed with Alex that businesspeople today must have an in-depth understanding of their consumers because those customers have a lot more power now than when I was a young man—which, believe it or not, was not that long ago! Most consumers then dealt with local retailers who enjoyed a near monopoly. Whether it was a department store, a hardware store, or an appliance store, most were located within a short distance from where customers lived. The selection was limited, and customers were almost entirely dependent on salespeople to provide information about a product or service. On the other hand, the merchants often knew their customers well and understood what would satisfy them. They often lived in the same town, had children who went to the same schools, and belonged to the same local organizations.

Lack of choice existed across other areas as well. When I was growing up in Southfield, a suburb of Detroit, we had three network TV stations to choose from instead of the hundreds of channels available on cable systems today (not to mention Netflix and Hulu and other streaming services). Audiences for popular major network shows approached 40 million nationwide, an almost unimaginable number today. If I wanted to find out what was happening in the world, I had the Detroit Free Press or the Detroit News newspapers available instead of hundreds of online editions of print newspapers from cities all over the world, a dozen cable news channels such as CNN and Fox News, and websites such as the Daily Beast, the Huffington Post, or Yahoo! News that are updated with breaking news throughout the day. My banking choices were basically limited to those branches in my neighborhood. Chase, Bank of America, and Wells Fargo were unknown entities, restricted by law from coming into my state. And when my parents wanted to book a vacation, which was usually not that far away, they went to the local travel agent. Being able to jump on the computer and scan hundreds of different vacation offers and instantaneously get information about dozens of discounted hotel room rates was as futuristic as an episode of Star Trek or a novel by Ray Bradbury.

As I finished my brief discussion of the immediate past, Alex had the same glazed-over look on his face I used to get when I heard my grandfather talking about World War II or iceboxes: that’s ancient history.

But this history is important to understand. Once, businesses had almost all of the power. When consumers walked through the door, they were almost captives. Businesses had the perennial real estate advantage (location, location, location) and almost all the information about products, services, and competitors.

Not anymore.

The balance of power—especially the balance of power in terms of information and choice—has dramatically shifted over time toward the consumer. This shift toward consumer power is a phenomenon I call Accelerated Darwinism—business survival of the fittest at breakneck speed.

Accelerated Darwinism is the result of a number of factors, but the communications and technological revolutions are, by far, the most important reasons behind the rapid pace of change and the shift in the balance of power between the consumer and business. It has resulted in the rise of what I call the Super Consumer.

Today’s consumers have amazing, superhero-like capabilities. They can:

  • Clone themselves by shopping in five stores at once through the use of multitabbed browsing or by utilizing multiple channels (shopping in a store while using an iPhone to browse other retailers).
  • Speak with a very loud voice where potentially hundreds, thousands, and sometimes even more than that can hear them express their feelings about their consumer experiences, good or bad, with any company by posting on Facebook and Twitter.
  • Have incredible range of hearing where they can listen to friends’ opinions on Facebook, Twitter, and e-mail and to millions of voices via social media rants and recommendations such as those on Yelp.
  • Have genius-level intelligence, often possessing as much or even more knowledge than a company’s employees by utilizing the web for customer-generated product reviews, detail specifications, and the competitive landscape.

Yesterday’s shoppers for a television walked through a single store, glanced at a few different models, possibly asked a salesperson for help, and made a decision. Today’s shoppers may do a little research online, cruise the aisles of a store to look at products in person, check online reviews with an iPhone, and then visit an online retailer to complete their shopping process—all the while comfortable and secure in the knowledge that if this website doesn’t have what they need, they can sprint off to dozens of other online retailers with a few clicks. And if their experience was great (or was horrible), they are likely to tell others on Facebook or Twitter what their shopping experience was like.

Because the customer’s voice is incredibly loud and because so many people can hear it, measuring the customer experience (and word-of-mouth) is a vastly more complicated phenomenon than it was even a few years ago, let alone what it was 20 years ago or for my parents and grandparents. The impact of the satisfaction with the customer experience is magnified by a factor of thousands and cannot be ignored anymore by businesses.

It is impossible for anyone to predict the new communication tools, much less which of those tools will capture popular imagination and be adopted by millions of users (Google glasses, anyone?). What we can guarantee is that any new tools will be broader and faster than the ones consumers use today. As a result, Super Consumers will share ideas and opinions with more people at a faster speed, and they will also listen to the opinions of many more people at a faster speed. This magnified voice of the customer will grow only louder. And the only way to keep up with these developments is to utilize increasingly sophisticated analytics to understand what customers want and allocate resources accordingly.

Alex nodded along—whether it was the force of my argument or the excellence of the local beer, I’m not sure. But when I finished, he asked, “So how do you recommend Ideal with my problems with NPS?”

I smiled and said, “You’ll just have to wait until tomorrow morning. But I have been having conversations like the one we just had for the past five years. Your concerns and issues are exactly what led my colleagues and me to work on developing WoMI. I am eager to hear what you and others think about what I say.”

The next day, after my speech, I did talk with John, Anna, and Alex, and each seemed to have learned something and to be eager to understand more about WoMI and the Customer Experience Measurement Ecosystem. And each seemed to clearly understand my ideas on how to evolve NPS, which is the focus of the next two chapters.


Buy Innovating Analytics at Amazon Buy Innovating Analytics at Barnes& Noble Buy Innovating Analytics at 800 CEO Read Buy Innovating Analytics at iTunes