CSP Happenings





Tagged: segmentation

The Dangers of Overlooking Low-Value Customers

March 23, 2017

One way to segment your customers is by their lifetime value. Compared to many other measurable customer attributes, lifetime value is the kind of big-picture description that can be difficult to observe or estimate at a glance. But it’s also one of the most valuable pieces of information you can have about your customers.

Customer lifetime value is a prediction of how profitable your relationship with a given customer will be over time. Lifetime value can be calculated a number of different ways, from simple formulas to complex equations. Some of the factors that go into this calculation include how long you expect the customer to stay a customer; how much money that customer tends to spend with you, and how often; what it costs to keep that customer loyal; and the average rate of churn throughout your customer base.

Essentially, a lifetime value measurement boils down your relationship with a customer to a dollar amount. But the benefit of is not just quantitative: it influences businesses to prioritize the long-term maintenance of customer loyalty, compared to more expensive efforts like customer acquisition.

Using Customer Lifetime Value for Segmentation

customer segmentation based on lifetime customer valueOnce you’ve predicted the lifetime value of each customer, you can then group them into tiers, from most to least valuable. The most valuable customers are those who shop with you frequently, generate the most profit, and are most likely to stay. The least valuable customers may be new or casual shoppers who split their attention and money between you and your competitors. In the middle are the rest: regular, if not devoted, customers who don’t have the most to offer you, but don’t cost you much to keep, either. These are customers who could possibly be influenced toward more value if tended correctly – and if not, may slip down to the bottom tier.

For the highest third, the ones you can’t ask much more of, the goal is to make sure they stay. For the middle third, you can try to grow their value by appealing to them with additional services or products, or offering loyalty rewards, like discounts.

But what about the lowest tier – the one that experiences the most churn? Do they have anything to offer? Would you be better off without them?

Don’t Write Off Your Low-Value Customers  

Once you’ve singled out your most valuable customers, it’s only natural to want to gravitate in their direction – to reward their loyalty with perks, to provide them the best service, and to otherwise do everything in your power to keep them around and keep them spending. All this effort is still less costly than investing the time, energy, and budget to convert lower-value customers up the ladder.

All of that would seem to justify prioritizing your top tier. But do so at your own peril. Every customer’s “lifetime” with your business has to start somewhere, and many of them start in that lowest tier. It’s rare to simply acquire a high-value customer out of the gate: they must be nurtured, and this tier is where that relationship-building has the most impact.

Converting an existing low-value customer into a higher-value one is still less expensive than acquiring a new one, with unknown value. Despite the investment they demand, it’s easier to see a customer move up the value ladder, while the ones at the top are not terribly at risk of slipping back down (unless you really mess up). Besides, if you don’t devote attention to the least committed customers, chances are that your competitors would be more than happy to take them off your hands.

The easiest way to make your low-value customers into VIPs is to treat them like VIPs.

No matter how many invisible dollar signs hang over their heads, every interaction with your business and brand should make them feel valued and respected. They can also be a valuable source of insights and intelligence. Don’t be afraid to ask them directly: What would make you shop here more? How can we best serve you? Voice of the Customer research is your friend, especially among this group. What applies to some can likely be used to woo others.

Of course, some customers will just never be converted and end up taking up more resources than they’ll ever be worth to you. Customer divestment was once practically an anomaly, but these days, some companies see it as a smarter move than keeping low-value customers on the books. (Read more: The Right Way to Manage Unprofitable Customers on HBR.org – though we at CSP don’t necessarily stand behind that headline.)

Bottom line: Give each customer the attentive and inviting customer experience they deserve, and watch the overall value of your customer relationships grow.


You may also want to read:

Customer Segmentation Pitfalls and Potholes

March 9, 2017

Customer segmentation can be an immensely useful tool in getting actionable insights from your customer research. From those insights, you can devise strategies to improve the customer experience, because you have a more specific understanding of what customers want. But segmentation is far from simple.

To get the most out of it, you need to understand a few things about the art and science of conducting customer research. That’s just what CSP has been doing for more than thirty years, so we thought we’d share some of our pointers.

What Can Go Wrong with Customer Segmentation

customer segmentation

Methodology Mishaps

What does your business have in common with the Large Hadron Collider – the massive facility in Switzerland that smashes atoms together to better understand physics? You both rely on the Scientific Method. Or at least, you should (and they certainly should).

Any good research, whether studying customers or plants or animals or atoms, is based on these standards, which have been the guiding principles of science since the mid-1700s. To get good results out of your research, your methods must be scientifically sound, unbiased, and verifiable.

Research is not just conducted, but designed. That means knowing how to create a sound and testable hypothesis, conducting the right kind of ‘experiment’ to test it, and verifying your results with the proper vigor. Get any of these parts wrong, and the rest unravels from there.

Contaminated Sample

Sometimes research starts from scratch, but often, it relies on parsing data you already have on hand. That might include one or more customer databases or Customer Relationship Management (CRM) tools. These databases must be meticulously maintained so as to avoid contaminating the results. Examples of database disruptors include duplicate entries, incomplete entries, “dead” entries (meaning, invalid or out-of-date information, such as dead email addresses), and false categorization.

A slip-up here or there may seem like not a big deal, but it can lead to disasters in customer communication. For example, in 2011, the New York Times erroneously sent out a special discount offer to a small list of 300 recent ex-subscribers to entice them back – except that it was delivered to 8 million contacts, including many current subscribers who suddenly became aware of a discount they were not being offered. Things like this can happen when database entries are not correctly or clearly identified and grouped.

You Know What They Say About Assumptions…

Everyone has conscious and unconscious biases and makes assumptions based on those biases – it’s only human, and it’s rarely malicious. But such assumptions, no matter how logical or benign, can still affect the viability of research results and the value you get out of them.

A good example of where you see this happening is in discussions of the different generations – Boomers, Gen X, Millennials, and so forth. Many sweeping generalizations have been made about each group, some supported by sound research, and others just created by socialization. Eye-grabbing headlines and op-eds easily filter through to form your beliefs about these potential customer segments.

When that happens, you are more prone to leaping to the wrong conclusion. Don’t assume that seniors don’t use mobile banking because they’re technologically illiterate, or that lower-income customers don’t have smartphones. Any conclusions derived from research must be supported by that research.

It Pays to Have an Expert on Your Side

Done well, customer segmentation can lead you to valuable insights and an improved customer experience. Done poorly, it can just as easily lead you astray, or not lead you anywhere at all. If it were easy, businesses like CSP wouldn’t need to exist – but luckily, we do. To learn more about how we guide businesses in using their data to provide stellar customer service, contact us


You may also be interested in:

SOURCES

New York Times email mishap
Unsupported assumptions

Localizing Customer Service & Customer Experience

February 22, 2017

Brand standardization used to be a priority, but retail businesses are increasingly foregoing out-of-the-box functionality and appeal in favor of individual touches based on location – a practice referred to as localization. There’s even a name for the niche branch of customer research that enables localization. Geodemography is defined by the Business Dictionary as the “process of analyzing survey data of a specific geographical area to profile economic and demographic characteristics of [the] population living there.”

Localization might mean using existing architecture instead of building another replica of your standard store model. (Some communities even enforce this in an effort to preserve local culture and history.) It might mean offering special discounts to employees of some of the area’s largest employers. Local sports team sponsorships, neighborhood events, and even high-tech tactics for garnering positive reviews for Google Maps and Yelp are all part of localized strategies.

Localization is not just applicable to marketing, though. Or really, it is, so long as you realize that the customer experience is a critical component of marketing.

localizing the customer experience
Why does localization matter? Shouldn’t we be guaranteeing the same customer experience no matter which of our locations customers walk into or call?

Well, yes and no. Yes, you should be guaranteeing the same quality of customer service. And a standard of familiarity is customer-friendly, too. You don’t want customers who are confused about where to find things or whom to talk to.

But providing a superior customer service is often about going the extra mile. That means anticipating customers’ needs and wants before they make contact with you. It requires knowing your customers well enough that you can tailor their experience specifically to them.

Geography is as much part of customers’ identity as other vital demographic statistics like age, sex, and income. It’s intrinsically linked to other identifiers, from socioeconomic status to school spirit, but it can also transcend those identifiers as a unifying factor. We are all in this (town) together.

Locality lends itself to in-jokes – you’re clearly not from around here if you don’t know that _____ serves the best pizza/wings/happy-hour nachos, or if you’ve never taken a date to _____. Whether it’s through hometown pride or well-intentioned humor, when local businesses participate in the customs of their surrounding communities, patrons and passerby alike will notice.

So when you’re designing your customer experience down to the last detail, that should include details specific to the locations of your branches. When you combine local knowledge with Voice of the Customer research, you create a customer experience that, literally, can’t be duplicated.


You may also want to read:

Customer Segmentation in the Big Data Age: Where Banks Find Value

February 8, 2017

Customer segmentation helps banks get to know their customers on a more granular level. Segmentation reveals specific intelligence that could otherwise be obscured by the sheer volume of data. These insights, in turn, inform messaging strategies for marketing and customer service strategies. Segmentation can also help banks better understand the customer lifecycle and predict customer behavior.

Examples of common customer segmentation criteria:
  • Customer value – How many products & services customers purchase and what kind of revenue that generates for the bank – past, current, and predicted for the future
  • Demographics – Age, geography, gender, generation (e.g. Millennials and Baby Boomers), income level, marital status, and other “vital statistics”
  • Life stage – Slightly different from age, focused instead on customers’ journeys through various milestones and markers; for example, graduating college or starting a family
  • Attitude – Customers’ subjective stances on things like the financial industry as a whole, online and mobile banking, the economy, and their satisfaction with their bank
  • Behavior – Interactions and transactions between customers and their bank, which channels they use and how often, and which products they adopt

Similar criteria can be applied to banks’ business customers – profitability, number of employees, “life” stage (start-up, established, legacy), and so forth.

These are the traditional ways that customers have been segmented for decades. However, relying just on these categories is not going to yield many actionable insights.

In the age of Big Data, you sometimes have to think small. The real power of segmentation is not the quantity of data you can collect – which, with today’s technology and methods, is virtually infinite. It’s in the ability to drill down to the information that actually teaches you something about your customers.

Often it’s not the segments themselves, but where they overlap, where you’ll find the most valuable intelligence.

customer segmentationSome examples: unmarried, home-owning, degree-holding women under 45; middle-income married parents of high-school-age children in a particular school district; and minority Millennials who are starting their own digitally driven businesses.  Any of these micro-segments may prove valuable customer niches for banks to prioritize. But first, you have to conceive of their existence. Second, ask the right questions. And third, conduct the relevant research to answer those questions.

To understand how this can come in handy for banks, just think about the sometimes bizarre categories that show up in your Netflix queue based on what you’ve been watching lately. Vintage sci-fi with a strong female lead? Critically acclaimed British nature documentaries? Criminal investigation murder mysteries based on books? The more they know your tastes, the more likely you are to keep using their service based on their recommendations.

The options for how segments can overlap are nearly limitless.

Nearly. There is a bell curve to the usefulness of segmentation. Too broad, and the results are less than insightful. Too narrow, and the value of the insights gained will have minimal bottom-line impact.

This is where it helps to have experienced data scientists on your side. The purpose and advantages of segmentation are easy to enough to grasp, but the farther you get into analytic methodology, the more highly technical it becomes, and the more you need to understand about mathematical models and formulas. If things like our guide to data visualization make your eyes glaze over, chances are that the nuances of segmentation will put you right to sleep, too.

But you’re in luck, because CSP’s customer experience & research experts are passionate about getting you the insights you need out of the wealth of data we can gather. So if you are interested in getting to know your customers down to the niche level that segmentation empowers, give John Berigan a call at (800) 841-7954 ext. 101 or contact us by email to start a conversation.


More articles on using data to your advantage: