Big Data and Customer Metrics: Time for a Change?
Big Data is no panacea, says the author of "Innovating Analytics." Focusing on the wrong data or not taking context into account can lead to bad decisions.
Big Data is often described by citing three of its characteristics: volume, velocity and variety. While many people believe focusing on those three things can help enterprises make better decisions, just the opposite may be true, says Larry Freed, CEO of customer analytics provider ForeSee.
"Big Data has incredible potential, but by itself it isn't going to solve any problems," said Freed, author of the new book "Innovating Analytics," which focuses on using data to measure, manage and improve customer experience. "More data isn't necessarily better data. Do you have the right data and the right analytics to go with it? That is where it can be really powerful."
Freed believes variety presents the greatest challenge – and the greatest opportunity – in dealing with Big Data. Most companies focus solely on behavioral data, but Freed said it is not always reliable in today's age of multi-device, multi-channel consumers.
"If I am an anonymous visitor to a website on my tablet, my home PC and my work PC, companies are not going to have the ability to recognize that I am the same person," he said. "Companies are gathering enormous amounts of digital data, but it can accelerate some bad practices."
Companies may also focus on business goals that do not mesh with the desires of their customers. For example, a consumer may open an online banking application to view mortgage rates and then abandon the app fairly quickly. But that may simply be a logical "part of the process" for a consumer who chooses to visit a branch location for a face-to-face meeting with a loan officer, Freed said. "It's hard to interpret without context. You can get lot of false positives or negatives -- and may make bad decisions because of it."
Companies should supplement behavioral data with voice-of-the-customer (VoC) data, often gathered via traditional tools such as customer surveys, Freed suggested. Yet many of those tools need to be tweaked to better understand today's consumers, he said.
Considering Customer Metrics
For instance, Freed said, the well-known Net Promoter Score (NPS) customer loyalty metric "has outlived its usefulness" and can be misleading. The NPS is based on the idea that customers categorized as promoters are better customers than those categorized as passive, and passive customers are better than detractors. A customer is deemed a promoter, passive or a detractor based on his or her response to the question: "How likely are you to recommend our company/product/service to your friends and colleagues?" Consumers are often asked to use a scale from one to 10 to indicate their answers.
Freed said research undertaken by ForeSee showed that NPS scores tend to vastly overstate the number of detractors. "When we looked at detractors, we found a large portion of them are actually satisfied and loyal and not prone to negative word of mouth like you might expect. The analytics are not getting to the right level of detail."
ForeSee has developed the Word of Mouth Index (WoMI), a metric Freed calls "the next generation NPS." While it builds on the NPS, a key difference is an added question: "How likely would you be to discourage others from doing business with company?"
"Not likely to recommend is not the same as negative word of mouth," Freed said. "We added another question to better understand the customer."
The WoMI's "true promoter" classification is much the same as a promoter classification on the NPS, but the WoMI's "true detractor" is "very likely to discourage" others from doing business with a company, and thus is more aggressively negative than the NPS detractor. "We subtract the percentage of true detractors from true promoters to get one number," Freed said. "It's a more accurate and precise measurement for negative word of mouth."
Negative word of mouth is more important than ever because of the "multiplying and magnifying capabilities" of social media like Facebook, Google + and Twitter, Freed said. In addition, he said, analytics can help companies "systematically identify the common characteristics of true detractors and determine what they should do," perhaps tweaking their pricing, product or marketing strategies.
Big Picture, Not (Necessarily) Big Data
"Maybe you are marketing in the wrong places or to the wrong people, or you have one product line that isn't doing well, or service isn't good in a particular geographic region," Freed said. "Data analysis gives you a chance to identify those kinds of issues and solve them."
Freed said companies should collect and analyze behavioral data (examining customer actions), feedback (directly gathered through Web forms and surveys and through social channels), and voice-of-the-customer data.
Big Data tends to focus more on behavioral data and feedback, and not as much on VoC, Freed said. But VoC should not be neglected, he cautioned. "The VoC measurement is more representative of your base than feedback and more forward looking than behavioral data."
A key takeaway is that "no one number tells you all you need to know," he said. "Positive word of mouth, negative word of mouth, retention, loyalty: Those are all driven by satisfaction, which is driven by a series of experiences. You really need to measure all of those to not only know how you are doing but where to invest and take action so you can do better."
Ann All is the editor of eSecurity Planet and Enterprise Apps Today. She has covered business and technology for more than a decade, writing about everything from business intelligence to virtualization.