PDF File: 

CRM5_wp_tanner.pdf

Baylor University and BPT Partners, LLC

Many CRM professionals talk about a 360-degree view or one version of the customer truth.Here’s how to create a data collection strategy to realize that view.

The 360-degree view of the customer is one of the most powerful concepts in CRM. But what does that mean? And how do firms actually create that view? Or do they? From what we see described as CRM strategy, and examples given of best-in-class CRM implementations, it is clear that few companies actually accomplish a 360-degree view. Why? Because the data are in different places, owned by different departments, in different formats, and collected via many different means (direct sales, email, Web sites, etc.). But even when we solve those problems, we don’t have a 360-degree view. The biggest limitation is what we thought was important in the past.

Behavior Is What Counts

Most companies believe that behavior is all that counts; they want to understand what their buyer buys. As a result, many CRM strategies end up limited to stimulus-response marketing, as in “customers who bought X also bought Y.” There is no doubt that there is value in peer recommendations based on market basket analysis, but there is also no doubt that there is no relationship in that communication. Because we don’t know why the customer ordered the original product, we don’t have much understanding of what else would fit that particular customer’s situation.

Behavior is all that counts when we are concerned only with making a sale. If the goal is to pump up the value of the current transaction by cross-selling, upselling, or full-line selling, then behavioral data is all that we need.

If, however, our goal is to segment to form a relationship with the customer, then we need more than behavioral data. We need motivational data, specifically in terms of the customer’s immediate need, in order to understand why the customer engages in that behavior. We want to create more pertinent offers and to position existing offers more effectively. To some extent, marketing researchers have been collecting this data for a long time; we just haven’t been using it for CRM.

We also need relational data. What is the desired form of relationship, from the customer’s point of view? For example, Bob Cohen, former product manager of M&Ms, used to ask, “How does this CRM stuff apply to M&Ms? How can I make use of this?” He wondered about whether Masterfoods could create relationships with individuals in a meaningful way. Many people have deep relationships or affiliation with M&Ms. As Cohen says, “Oftentimes, people feel compelled to share with me their M&M story, such as eating them as comfort food when traveling in other countries.” These customers already had a deep relationship with the product. Masterfoods’ challenge was not in creating the relationship, but in leveraging the relationship.

In fact, one of the earliest theoretical approaches to relationship marketing was developed by Bob Dwyer and Seja Oh at the University of Cincinnati. They recognized that there are relationships where the motivation for a relationship comes primarily from the seller, but there are other relationships, like that of M&Ms, where the customer drives the relationship. We seem to have forgotten, except when we want to cut costs, that there are customers who are willing to manage the relationship and will keep it a close one. Let’s examine relational data, how to collect it, and what to do with it.

Relational Data

In our personal lives, we have four types of relationships – family, friends, acquaintances and strangers. Within each type of relationship, there are differences: How I treat my spouse is different from how I treat my child. But when we talk about those four sets of relationships, people know what we’re talking about. These relationships are defined by the level of commitment, the level of knowledge both parties have, the degree to which they share mutual goals and values, and the degree of contact they maintain.

Most companies assume that they have a linear set of relationships: great customers, good customers, bad customers and non-customers. The reality is that customers don’t think of their suppliers as great suppliers, good suppliers, bad suppliers and non-suppliers. They say, “This is my financial adviser,” or “This is my car dealer.” Sometimes they have multiple vendors for the same general needs, with different roles for each vendor. Customers have a set of expectations for each role, just as we have expectations for our spouse and our child.

Consider collecting expectations for more than just service levels. For example, very good customers of a financial institution may have expectation levels for service that relate to payment terms, product availability, waiting time on the phone, and that sort of performance metric. At the same time, however, we need to understand their expectations for commitment, contact, knowledge, values and goals. There are probably other dimensions that determine relationship form as well, but to assume that customers tier suppliers in the same linear fashion as companies tier their customers avoids benefiting from relationship quality.

In a salesforce-dominated environment, the salesperson makes the adaptation to the customer’s relationship expectations. When the customer interacts with us via technology, or through multiple channels, then we need to create mechanisms for understanding relationship expectations. For example, we have touch-tone menus that take a caller to different levels – why not have touch-tone surveys while the caller is waiting? Ask one or two questions and collect the data. You know who it is (they have already entered their account number), so using the call to collect some information is not too difficult. Similarly, when a customer is accessing account information via the Web, use a pop-up to ask a few questions and begin to collect data over time. Make sure, however, that they recognize a benefit for participating.

Relational data is one important type of data. It helps us understand why someone wants, or does not want, a relationship with us. Relational data, however, is still only part of the total picture. We also need to understand motivational data, and if we’re in the B2C environment, we need to understand lifestyles, too.

Motivational Data

Traditional marketing research is great for understanding motivations. We have a number of methods that help infer motivations within samples. But our goal is not to sample our market and determine the strongest motivator for the entire market. Instead, we’re trying to find out what motivates our customer within a specific set of transactions.

As an example, Collin Street Bakery in Corsicana, Texas, wants to segment its gift buyers according to the importance of the gift. Is this a duty gift, a status gift, or a loving gift? Research indicates that these are the three dominant gift-buying motivations – by offering gifts that qualitative research has defined as one of the three, the transactional data helps Collin Street Bakery infer motivation.

Similarly, most companies only record positive transactions. Few record negative transactions. We know who responded, but do we use the information as to who did not, and do we model or segment based on the “no’s” as well as the “yes’s”? From that, we can begin to infer motivation.

Even within positive transaction data, we lose information if the transaction is all we record. For example, if I buy a baby book on Amazon.com, does that mean I am having a child, a grandchild, a niece or have a duty to buy someone a gift? Wouldn’t knowing that help Amazon? The data are not only easy to collect, but value could be added by offering gift wrap, or a hand-written card, or direct shipping. What happens, though, is that even when these additional services are offered, the data aren’t retained. One friend described baby product emails she subsequently received, even though she was just buying someone a gift.

E-Rewards, an online marketing firm, does a great job of collecting some motivational data by asking its customers to rate the relevance and importance of each offer. Even when I say “no,” I can tell E-Rewards at least some information. The firm could probably go further and ask why, but the ongoing collection of such data helps E-Rewards pinpoint offers to its customers much more effectively.

Every time we talk with a customer, whether through technology or interpersonally, we have an opportunity to learn more about what makes that customer tick. Just a little information gathered each time will help us to understand their motivations.

Lifestyle is an important set of motivations. Lifestyle is a combination of a consumer’s self-image, the activities she participates in, the values she holds, her socio-economic status, and more. For example, the American Paint Horse Association (APHA) found that its members also like to hunt and fish. So APHA began to advertise in hunting and fishing magazines to attract new members. More importantly, APHA began designing offers with partners in hunting and fishing in order to bring greater value to its members.

Collin Street Bakery’s signature product is fruitcake. It would be helpful to know if the fruitcake was purchased in expectation of serving at a party, while entertaining houseguests, for personal consumption, or as a gift. Collin Street Bakery doesn’t have to only say, “People who bought our fruitcake also enjoyed our pumpkin pie.” The basket analysis is more powerful; it allows the bakery to make appropriate offers such as, “When you entertain this holiday season, consider serving these specialty cookies!” More importantly, it can begin to identify those who entertain often versus those who have family or other houseguests often, two important lifestyle characteristics.

Understanding lifestyle isn’t just about understanding product use. Amazon may be able to figure out that someone is interested in horses if only horse books are purchased. Buying murder mysteries, though, doesn’t mean the customer’s hobby is killing people. Furthermore, the benefit in building lifestyle data is not just about creating an individualized offer. The challenge is to find a way to utilize intelligent computerized systems that identify the corresponding lifestyle of each customer when, like Amazon, the company has millions of customers. Lifestyle data, though, is needed to create communications that are meaningful.

For example, one electronics retailer managed to capture aging baby boomers by positioning direct mail pieces toward appropriate life events. As customer Carole Mercer said, “When they showed someone like me videotaping their grandchild, I knew it was a product I would use.” The offer itself was not “lifestylized,” but the promotional piece was. While age is part of a person’s lifestyle, consider the value of even deeper lifestyle knowledge that can be used to tailor the presentation of an offer. With deep lifestyle knowledge, the message can be tailored to make the relationship stronger.

Collecting The Data

Our research indicates that people interact with marketing communications because they expect a benefit from that interaction. Similarly, they expect to benefit from giving you data, particularly if the data collection process is onerous. Yet, when we learn about our friends, acquaintances, and family, usually that process is far from onerous. In fact, it is natural.

If there’s one hidden principle so far, it is this: collect data as part of the conversation to make it as easy as possible on the customer. There should be data collection and entry opportunities when a customer service rep takes a call, for example. Rather than turn the call over as quickly as possible, consider asking two questions that could be answered easily and put into the system. Even text answers can be categorized and used in sophisticated modeling tools to create a more complete picture of your customer.

Another principle for data collection is to think long term. What do you know about your customers now and what would you like to know? Don’t try to answer it all at once. Just as you and your spouse have changed as individuals over time, thus changing your relationship, your customer changes too. A single snapshot of motivational or lifestyle data is just that, one snapshot. Give the customer a few years and he won’t be like that snapshot any more than you are the same person you were in your high school picture. Instead, think about what you want to learn and structure interactions so that you can continuously collect naturally occurring data.

Finally, think about how you plan to use it. Every one of us has designed the perfect marketing research study only to realize afterward that if we had worded one question a little differently, the data would have been much more useful. Additionally, there is always a cost associated with data. There is a cost to you and a cost to the customer. Make sure the use justifies the cost. As you use your data in creating models of your customers, identify and correct the shortcomings.

When we use ongoing, naturally occurring dialogue to learn about our customers’ lifestyles, motivations and expectations, we are much closer to that 360-degree view.