No Comments Written by Robert on July 4, 2008 – 9:55 pm

CRM is not a mass program, it’s a best-customer program. If you run it like a mass media program you’ll lose money and effectiveness. CRM is a one-to-one medium. It’s about two-way communication that says to the customer “they know me and want to find out more about me.”

Mass advertising tries to make every person who sees the ad happy and feel good about your company. CRM is about evaluating customer potential and spending more on those that mean more to you. CRM is all about best customers. For your CRM program to succeed, you need to identify clusters or segments of your customer base that will respond to very targeted communications at greater rates than a mass audience. If you try to turn your CRM into a mass marketing program you’ll find that your expenses outpace your revenues.

The cost per thousand reached in most mass market vehicles is a fraction of the cost of meaningfully interacting with a CRM customer via mail or events. The one place in which the cost is extremely low is Internet contact. However, the time involved in really learning and dissecting segment needs and wants and then translating these into products and communication programs is very extensive.

At Famous Footwear, we have one target market segment called the Fashion Value Mom (a woman 35 to 54 years of age, with children, who loves shoe purchasing, is a spontaneous shoe purchaser, and purchases for the whole family but buys more shoes for herself especially dress and casual shoes).

This Fashion Value target market segment accounts for 32 percent of all purchasers and 46 percent of all pairs of shoes purchased. Additionally, we have nine segments under this target market segment based on purchase behavior.

For example, 25 percent of the target market falls into the “whole store” segment those that make purchases across the whole store. There’s a “kids” segment that primarily uses Famous Footwear for kids’ purchases, a “fashion athletic” segment, a “performance athletic” segment, a “fashion dress” segment, and a “casual” segment, along with three other segments.

These segments form the foundation of our CRM program and we use this segmentation to merchandise and communicate to these segments on a regular basis. These segments make up our best customer program and account for 25 percent of all total customers and 50 percent of the organization’s sales. Each segment receives different communication based on the products they purchase, the offers to which they are most responsive, and the receptivity they have to move into other categories of purchases. It’s easy to get lost in the details of CRM.

• Don’t oversegment. In the extreme, you could end up with each individual customer being a segment. Oversegmentation is how CRM programs end up with negative return on investment. Find meaningful segments but remember, the software today makes it easy to create segments that are so small and have such minute differences that segmentation simply becomes an academic exercise.

• Invest in the segments that yield the biggest dividends. For example, at Famous Footwear, three of the nine CRM segments make up 60 percent of the CRM database and also have the largest lifetime value. It is on these three segments that Famous Footwear spends the majority of its CRM dollars.

• Don’t overtest. You can test, test, and test some more without really making any fundamental changes. And don’t get stuck in analysis paralysis, in which you test the same ideas again and again without any decisions on improvements. It’s an easy discipline in which to simply “do tasks” and appear to be very, very busy (and in need of more and more money for implementation).

CRM is all about measurement, but measurement is important only if you test what is meaningful and act on what you learn. There are a million things you could be testing. Test those with the largest potential to significantly improve your business.

CRM technology and software allows for the ongoing testing of ideas. The trick is to understand the metrics that are most important to the organization and then to measure against these.

The following are examples of things a company might prioritize for testing.

• What’s the incremental response time from new members receiving their first contact?

• What combinations of incentive and product inserts and timing (inserts combined with incentives or as a follow-up) work best to promote redemption and total response?

• What recovery (getting back lapsed customers) incentive works best and is most profitable?

• What creative formats work best?

• What segments are the best predictors of incremental response to various incentives?

• What pure product pieces drive the best response for different segments?

You should have a yearly test plan. Don’t just randomly test things, but concentrate on an area and do enough tests so that you can find the right answers and move on to other areas. Remember, you can only test one variable at a time, so you often have to run multiple tests of similar ideas to separate out what works best.

Make sure to hold out control populations that allow you to statistically prove that the tests perform below, the same, or above the control or the group of customers that did not receive the test communication.

The biggest challenge is to act upon what the tests show you. We’d suggest a standing meeting of cross-functional leaders who meet to review test results and determine the action to be taken in the future.