“Those who insist that every [customer] is created equal merely ensure that ‘the worst customers get better service than the best,” Bob Sabath, a discipline expert in supply chain management with Trissential, told the editorial staff at SupplyChainBrain. “Companies need to be thinking about new and creative strategies for serving their most valuable customers.” [“Breaking the Rules of Supply-Chain Management,” SupplyChainBrain, 25 July 2013] This is true whether you are discussing current customers or potential new customers. The most desirable customers are generally labeled high value customers. Michael Greenberg, vice president of marketing for Loyalty Lab, writes, “Everyone has customers who are more valuable than others. But why do you have them? Put another way, why did they choose to become a high-value customer?” [“Why Do You Have High-Value Customers?” Chief Marketer, 24 April 2006] Perhaps there is a more important question that needs to be asked first, “What exactly is a high value customer?”
Often the designation high value customer is given to an individual or company with a high customer lifetime value (CLV) — “the dollar value of a customer relationship, based on the present value of the projected future cash flows from the customer relationship. Customer lifetime value is an important concept in that it encourages firms to shift their focus from quarterly profits to the long-term health of their customer relationships.” [“Customer lifetime value,” Wikipedia] In order to determine which customers deserve the high value descriptor, Greenberg indicates that you must look “for variation between your high-value and low-value customers. This requires a fair amount of data concerning your customer spending, so that you can separate your customer base into a reasonable number of groups and still have enough response data to generate significant results.”
In today’s emerging era of big data, obtaining and analyzing data is becoming easier. Mark Zilling, Executive Vice President & partner at MeritDirect, claims, “The potential ways to use Big Data for customer segmentation are only limited by the imagination.” [“All Customers Are Not Created Equally in Value,” Direct Marketing News, 1 March 2013] He explains:
“Big Data takes many forms and can be aggregated from numerous sources. … Over the past few years, behavioral information has introduced yet another level of data sitting between the static demographics of compiled data and the proven buying characteristics of customer files. Today new technology provides the opportunity to link online activities, like email responses or website visits, to these large aggregated data sources, further enhancing the behavioral category known colloquially as ‘hand raisers’ or those displaying interest in specific products or services. Together these data sources integrate to create a Big Data platform that changes the landscape for marketers focusing on customer segmentation.”
Zilling offers three ways marketers can use Big Data to improve segmentation (i.e., identify potential high value customers from lower value customers). The first way is by “identifying probability of high-value, long-term customers at first order.” He writes:
“Marketers typically focus on new customers with the intent to convert the highest number possible into repeat purchasers. While this is a noble endeavor, it’s simply not practical. All new customers are not equal in regards to their future potential and the traditional tools are not adequate to segment greater potential customers from those of lesser value. In fact, new customers all have the same recency and frequency, leaving only the size of first purchase (monetary) and the product purchased as the characteristics on which to base segmentation decisions. By integrating Big Data, additional attributes can contribute to reorder models that have proven highly predictive in not only determining which customers are more likely to repeat purchase, but also how much they’re likely to purchase and what product or category it will most likely be in the next 12 months. This information is useful in determining what frequency and channel of communication to use in converting the site, and what offer or messaging strategy to employ. It can also help define which customers to send to the sales force well before they’ve self-selected or before a competitor has a chance to pry them away.”
Those all are excellent points. Greenberg notes that superficial data analysis is not sufficient. “It is likely you will find only a few differences in [the] responses to preferences, attributes, promotion, satisfaction, and so on [of your highest and second highest segments of customers]. You’ll have to get specific. Don’t ask about just three or four areas – ask about 30 or 40. You may find that only two or three show enough difference to warrant further study. The payoff comes from attention to the details, however, so looking at the differences in very specific areas can lead to a big payoff.” Zilling’s second recommended way to improve segmentation is to identify “higher-potential customers while they’re prospects.” Doing so, he writes, will save money. He explains:
“Big Data … can identify prospects who are not only more likely to respond and become a new customer, but of those prospects who are likely to become higher-value customers. This allows for a segmentation strategy that either avoids lower-value potential customers before spending money to acquire them, or tailors the acquisition and subsequent retention strategy to a lower-cost approach, creating a balance for the lower potential revenue and an acceptable ROI.”
Greenberg notes, “Your lower-value customers will come into play as you look at your competitors, which is the second area to examine. Study how your customers rate your competitors on the same attributes and what their preferences are. It’s very likely that your low-value customers are high-value customers somewhere else.” Zilling’s final recommendation is to link “website visits to marketing media” in order to recapture leads lost to e-commerce.” He continues:
That kind of analytics simply wasn’t possible a few years ago. Today, however, Zilling concludes, “The opportunities are as endless as the data itself.” Paul French agrees with Zilling. “Data lies at the heart of the modern sales success story,” he writes. [“How Can I Spot High Value Customers Faster?” Fliptop Blog, 31 May 2013] He also agrees with Greenberg that you can learn a lot by knowing why your low-value customer is someone else’s high-value customer.
“What makes a high value customer choose to become a high value customer? What steers their path from one-time purchasers to long-term business asset? If the question seems obvious, try pinpointing the answer. All sorts of resources go into marketing, but perhaps not as much towards identifying precise success and failure points, improving customer retention and over delivering where you have under promised. Within these walls are the secrets to effective customer service, where good management will safeguard the long-term success of your business. Are your low value customers a competitor’s high value customers? And why? Crack open a spreadsheet table and get specific. Run a warts-and-all assessment of customers’ preferences, satisfaction and attributes. But don’t stop there. Go deep. Look for as many as fifty areas where you can make things easier for people to love you.”
Obviously, the only way you are able to “go deep” is by using big data analytics. Much of the data that will be required for this deep dive is unstructured; which means that natural language processing is an essential capability if you are really going to understand the nuanced differences between high and low value customers. French concludes, “When you’ve got access to that data, you can identify the high value customers and their place in your sales funnel, leading to increased efficiency, customer satisfaction and ultimately business success.” John S. Parke, president and CEO of Leadership Synergies, LLC, agrees with the other pundits cited above. “Successful companies are using sophisticated analysis and deployment strategies to segment customers into categories such as ‘high value’ and ‘low effort’,” he writes. “This lets them allocate a disproportionate amount of resources to high-value customers, resulting in improved loyalty and higher revenues from customers who are good for business.” [“Not Every Customer Is King,” Executive Update, November 2004] He claims that “successful companies employ these simple concepts”:
- Not all customers are of equal value. Segment your customers and focus on those who will help deliver higher profit margins and make you more money.
- Loyal customers are good for business. A loyal customer costs less in terms of marketing, sales, and service. That saves money.
- Allocate a disproportionate amount of time, money, and resources to acquiring and keeping high-value customers. This is a good return on investment.
- Customers who want to form a business-to-business partnership instead of treating your organization like a commodity are more profitable over the long term. They tend to buy more and complain less, they collaborate on innovation and product development, and they keep switching costs high. (Simply put, switching cost is the amount it would cost a customer to move from your organization to your competitor. This cost is prohibitive because you already have a multifaceted, long-term relationship with the customer.)
Parke concludes, “Corporations need to make tough choices. If they’re smart, they’ll keep the good customers and let the competition have the low-value customers. This way they drive up their competition’s operating costs while driving down their competition’s profit margins.” The bottom line is that you can’t make smart choices if you don’t have access to the data and analysis necessary to provide decision makers with actionable insights.