In previous columns, I’ve discussed the benefits of using behavioral data and triggered email to improve campaign relevancy and, ultimately, response rates. Although past behavior is often the best predictor of future behavior, this month we’ll take a look at attitudinal analysis to create a customer segmentation model that’s representative of both behavior and attitudes.
Attitudinal Behavioral Data Largely Underutilized in Campaign Segmentation
A June 2004 Jupiter Research (a Jupitermedia Corp. division) executive survey asked about the data attributes used in segmentation schemes. Just 11 percent of respondents use pass-along (forward to a friend) rate. Most rely on typical recency, frequency, and monetary (RFM) schemes.
Though RFM analysis is undoubtedly effective and should be the basis for audience segmentation, marketers should consider it just a starting point. Beyond that, work to incorporate attitudinal data to augment what you already know about customer spending habits.
Use Satisfaction Survey Data in Audience Segmentation
The easiest way to discern clients’ attitudes about your products and services is to ask them. Though I’ve found satisfaction surveys are a widespread practice, most companies use that data only as a barometer to gauge clients’ overall sentiment.
For example, an April 2004 Jupiter Research executive survey found 71 percent of respondents use this data as an overall thumbs-up/thumbs-down barometer. Less than a quarter (22 percent) used that data as a segmentation attribute.
Many email service providers (ESPs) offer survey and polling tools embedded in the campaign management tool, making it relatively easy to collect this data and associate it with individual customer profiles.
Following are some real-world examples of how companies use satisfaction data:
- A high-tech consumer goods manufacture uses both satisfaction survey data and customer service call history to determine at-risk customer relationships. These customers are removed from the typical promotional campaign segments and populated in a satisfaction segment. This segment gets treated with campaigns that are much richer in editorial tone, including hints and tips on how to get the most out of the products. Promotional campaign frequency is reduced; these clients are instead presented with the support and informational content the attitudinal data indicates is more important to them.
Since adopting this formula, the company has experienced a 13 percent increase in customer satisfaction, which it’s found to be the best predictor of customer retention.
- A financial service institution uses surveys to identify at-risk customers. Customers who meet this requirement are segmented, and an email is sent to individual account representatives. They, in turn, call these customers. To offset the costs of these staff-intensive calls, the company moved its survey operations from an premium outsourced survey service bureau to a much more cost-effective hosted application. The application can trigger these call alerts based on predefined business rules.
Directional Pass-Along Data More Valuable With Behavioral Data
I mentioned the low number of companies using forward-to-a-friend-derived pass-along data. Though that number is low, we must also consider many companies have downplayed forward to a friend in marketing efforts in reaction to CAN-SPAM legislation. Although the law indicates a forward mechanism within a campaign must maintain the marketer is the sender, many companies and ESPs have developed mechanisms to ensure the address that receives the forwarded message is not on the opt-out suppression list.
If you go to such lengths to ensure a campaign’s forward-to-a-friend element is CAN-SPAM compliant, it provides a wonderful attitudinal segmentation opportunity. Here’s one that’s worked:
A large cosmetic company found some lapsed customers (determined by a lack of email and spending action) were actually forwarding messages promoting club membership. It resulted in about 5 percent of the company’s new monthly registrants. The company can now identify these brand champions and continue to mail to them, even when their click and spending behavior indicates they’re lapsed.
Often, we judge customers solely on their spending levels. Using pass-along behavioral data, we can identify customers who act as a valuable acquisition source even if their own spending doesn’t reflect that. To recognize these customers in spending-oriented RFM analysis schemes, put the average cost of customer acquisition on advocates’ profiles as a positive value to recognize their goodwill efforts. The tactic makes it easier to recognize them in monetary-oriented segmentation.
As with everything else, this approach requires some testing. Give it a try, and let me know how it goes.