TOO MANY LOYALTY marketers think they have to be all things to all customers; they design programs that are either too expensive or offer inadequate rewards. The far wiser course is to focus on the segments that will do them the most good, and not waste significant resources on the others.

Here’s how a cell-phone company created tailored programs for separate and unequal groups. The firm had many problems, including a high customer attrition rate, declining revenue per subscriber and many low-end customers. What’s more, its competition was changing rapidly due to price wars, new technology and alternative distribution channels.

The turnaround strategy was anything but easy. The company knew it had to reduce attrition, increase customers’ lifetime value and get a greater share of their spending. It also wanted to allocate its marketing budget more effectively.

To do this, the company conducted two studies through its service bureau:

• Segmentation and customer satisfaction

  Researchers found that the best customers perceived the least value and were least satisfied with the company’s service.

• Attrition

  Analysis showed that defectors expressed low satisfaction months before they decided to leave. It also revealed that the odds of defection quadrupled when a customer griped, and that those who were “very satisfied” with how problems were solved were unlikely to jump.

The next step was identifying potential defectors. So the company:

• Developed 68 different statistical models exploring all aspects of the attrition problem.

• Rated the models based on performance vs. a control group.

• Chose a neural network that could handle many variables for the final segmentation job.

The key findings of the neural-net model were:

• Two-thirds of all defections occurred within 15 months.

• About four out of 10 defections were preventable.

• 53% of preventable defections happened before the seventh month of a subscriber’s contract period.

The stroke of genius was to use quadrant analysis to create customer segments based on revenue and risk. This work indicated the segment where the company could get the most benefit from a loyalty program, and recommended that it design a program for this critical group of customers.

Based on the models, the company created a targeted rewards program focused on Group A, the high-risk types (those most likely to defect) with the highest current revenue. This segment represented only 13% of the entire customer base. Less costly rewards programs were set up for the other groups, some 87% of the customers. These strategies were intended to:

• Identify the four key customer segments, adjusting them as new ones arrived, old ones left, and existing customers moved up or down.

• Allocate marketing investment based on revenue and profit for each segment.

• Provide different strategies and rewards to each segment in the loyalty program, and within that sector offer individual loyalty rewards based on a customer’s life stage, needs and value.

• Provide enhanced services and proactive communications to the best customers in Group A.

• Use the models plus data on customers’ daily behavior to detect defection problems and resolve them before the customer headed for the door.

The company implemented the recommended rewards programs, creating a control group for each of the four segments to measure effectiveness. A year later, the firm found that:

• The programs generated a return on investment of $2.09 for every $1 invested.

• Attrition of those customers receiving the rewards communications in Group A was 1.27 points lower than those in a control group.

• Average revenue ($1,412) in the rewards test group was 5% higher than in the control group ($1,358).

• There was an increase of $19.6 million in annual sales to those 13,881 customers who were retained by the loyalty program (compared with a control group).

This last result is most impressive to me. The models were used to identify exactly how many customers did not defect as a result of the loyalty program.

Too often a loyalty program is launched and conducted without knowing how much good is being done. If this is so, how can any company prove if an effort works? The answer: By setting up a control group that doesn’t get the benefits. But this often is the most difficult part of any rewards program.

The control-group problem is illustrated by the recent experience of a master marketer. “To prove that the money spent on the program was paying off, we had a control group,” he said. “The company never had one before. It’s difficult to measure a program unless you have a control group.

“When we set up the control group we made sure that no store executives were in it, no board members and no employees. But we didn’t have a smart enough database to tell us who was the next-door neighbor and best friend of the president’s wife. She was in the control group.”

And the result?

“We got some angry calls from people who were in the control group,” the marketer said. “Of course, we didn’t tell them that they were in a control group. We just told them that there had been a terrible mistake, and shifted them to the test group. These people were tagged as the ‘out of control group.’ Others who were not qualified for the program called to complain about being left out. We explained that you had to spend so much to qualify. Most said OK, but for those who were adamant we made exceptions.”

A nagging question comes up when companies design loyalty programs: How large should the rewards be to spur changes in customer behavior?

One telephone company tried four rewards: a letter thanking the customer for his or her business; a similar letter offering 50 free minutes; and two similar letters pitching 100 and 200 free minutes. Which letter produced the desired behavior change at the least cost? The one without the minutes produced only very slight improvements. The three letters offering free minutes all worked equally well.

Conclusion: Giving a valuable reward works better than no reward at all. But if you are giving a gift, it really doesn’t matter how large the gift is. The phone company settled on 50 minutes as the most cost-efficient reward.

ARTHUR MIDDLETON HUGHES is a vice president and solutions architect at KnowledgeBase Marketing. He is the author of “Strategic Database Marketing, Third Edition” (McGraw-Hill).

Event marketing spending neared $171 billion in 2005, up an estimated 3% for the year, per PROMO research.

But the average event-marketing budget fell 12% in 2005 to $795,147, according to PROMO’s Event Marketing Trends Report survey.

That means more brands are using events, but keeping tighter reins on spending and driving hard for ROI. In fact, 96% of marketing execs use events in their mix, according to The George P. Johnson Co.’s annual survey, EventView ’05/’06 (formerly The Global Event Trends Survey).

Budgets are spare: 32% of marketers spend less than $50,000 a year on events; 15% spend between $50,000 and $100,000, and 19% spend $100,000 to $500,000, per PROMO’s 2006 Industry Trends Report. (Another 24% spent upwards of $500,000; 10% didn’t answer.)

“All brands are stepping closer to consumers,” says Brad Nierenberg, CEO of RedPeg Marketing. “Good event execution is just table stakes now.” So are results. Half (54%) of marketers measure event results by sales volume, per PROMO’s Trend Report. Half (50%) track headcount; 41% measure purchase intent; 38% track sales data; and 35% count Internet hits after an event.

Some shops query audiences as a research plus-up. “We’re pushing down to the grassroots with better service, and pulling up from the grassroots with better consumer insights,” Nierenberg says.

The push for higher ROI has marketers cutting street-team budgets. That pressures agencies to keep staffs lean — and makes well-trained, reliable street crews a hot commodity.

Most brands rely on buzz marketing to drive event traffic; that’s likely to grow as marketers co-opt pop culture (think film screenings and concerts) and recruit brand ambassadors. Watch as brands’ desire for entertainment content increasingly informs event planning.

“There’s so much competition for mind share — podcasts, social networking, socialized content via new media,” says Civic Marketing Group Co-CEO Stuart Ruderfer. “How can brands distinguish themselves? The only things working now are physical and digital.”

Many marketers are supersizing events, especially to reach young adults. “They’re taking over entire hotels, not just hosting a Spring Break event on the beach,” says Alloy Media + Marketing Executive VP Derek White.

Could that also mean the return of the big rig? GMR Marketing is building out four this spring, despite a shift to smaller, nimble vehicles the last few years. “It could be a coincidence as a few clients go all-out,” says GMR President Craig Connelly.

Watch events evolve through cell phones and social networking. Verizon Wireless gave away 10,000 tickets to its Grammy-week Fugees concert via street teams and text messages: Verizon Wireless subscribers who responded quickest got a “ticket” via MMS, then showed the bar code on their phone screens to get into the show. Phones (and the Internet) foster follow-up, too: “The ‘what’s next,’ continuing the conversation with consumers is built into nearly every program,” Connelly says.

Watch this year as event marketing segues into retail as more brands adopt “assisted selling” training models as brand experts talk with shoppers in the aisles — like the Science Diet experts who stroll through pet stores at peak traffic times.

SNAPSHOT 2005

Spending hit $171 billion, up 3%

Average budget fell 12% to less than $800,000

Buzz, cell phones drive traffic

Lean street teams stretch margins

E-mail marketing is under siege again, with new moves by spammers clouding the medium’s prospects as a trusted information resource for consumers.

Spamhaus, an e-mail monitoring group, reported in early February that software available now can send spam directly over providers’ mail servers rather than from hijacked PCs, thus bypassing the Internet service providers’ filter systems. According to the agency, America Online was reporting that more than 90% of its incoming spam came straight from the mail relays of other providers. And Spamhaus itself notes that at its current rate of increase, spam might constitute 95% of global e-mail traffic by the middle of this decade.

It’s a disturbing trend, given the ISPs’ role as a first line of defense against unauthorized or illegitimate e-mail. Authentication standards may help alleviate the problem, but so far those measures are being applied sporadically and unilaterally, with no sign of agreement on a single unified protocol. And when it does arrive, authentication alone may not clean up the spam pollution. A September 2004 study from security software firm CipherTrust found that spammers were adopting one leading protocol, Sender Policy Framework, even faster than legitimate e-mailers.

But spam is in the eye of the beholder, and e-mailers need to think Can Spam to get through the most important filter of all: the beleaguered recipient dealing with a cluttered e-mailbox, who may not even remember having given permission to be e-mailed in the first place.

This final gate can be tricky because the definition of

Andrew Wexler is the president of a business that relies on finding prospects online. His Boca Raton, FL, company, MoreVisibility.com helps e-commerce concerns position themselves on search engines to get maximum attention. Wexler does 20% of his prospecting through e-mail marketing. He’d like to be doing 50% to 70%. But there just aren’t enough qualified business-to-business names available.

Wexler’s experience is representative of B-to-B e-mail marketing. There aren’t enough lists to keep up with the demand. But that’s not stopping marketers like Wexler from testing e-mail campaigns and waiting for more files to appear.

In total, there are probably 100 different rentable B-to-B opt-in lists, containing perhaps 10 million names, experts say.

“It’s a tiny number, and they are not yet as niche-targeted as they need to be, but they will be soon,” observes Richard Baumer, president and CEO of VentureDirect Worldwide, New York.

List managers such as Pearl River, NY-based e-Post Direct and VentureDirect Worldwide handle a variety of lists that businesses have put on the market. But the largest and best-qualified names in the B-to-B sector are owned by large publishing concerns: IDG List Services, Framingham, MA; Cahners Business Lists, Des Plaines, IL; and Thomas Business Lists, New York. Thomas has two e-mail lists containing 700,000 names from its registries.

They cannot meet the demand. “I get calls daily from people who are looking for e-mail lists that are not in our niche,” reports Deb Goldstein, president of IDG List Services, whose company has 35 files, with about 800,000 names of subscribers to high-tech publications.

Wexler perseveres because the channel can be so effective. “The beauty of e-mail marketing is it’s immediate,” he says. “An e-mail can provide a link immediately online for them to respond to. From a marketing perspective, that’s tremendous.”

What’s more, business people are at ease with e-mail, and are a captive audience. “The business professional sits at a computer and deals with e-mail all day, compared with a consumer, who fields his e-mail once a week,” remarks Goldstein.

And business people generally respond to e-mail if they have requested to receive information from the mailer company (opted in), and if the information is useful and relevant to them.

Response rates have dipped in the past few months, but are still solid. As workplace e-mailboxes have become fuller with all sorts of promotions, “we don’t hit the super-highs like 30% to 40%,” says Denise Moser, director of data product development at Cahners. “The average campaign has maybe dropped 2% to 4%. But it’s still better than direct mailing.”

Of course, opt-out rates are higher than they are for paper mail. Last fall, as Cahners was beginning the process of appending e-mail addresses to its 200 controlled-circulation B-to-B subscriber lists, opt-out rates ranged from 2% to 7%. “Now, the overall opt out on a particular list is about in the 15% to 20% range. I think that’s about industry average,” adds Moser. Cahners will have completed the conversion process by mid-summer. Moser expects 1.8 million names from a variety of industrial and manufacturing publications to be available then.

Another issue is expense. The average price of a B-to-B file is $300. Mailers haven’t complained much yet because names are scarce, but “big mailers – who are sending millions and millions of names – are already negotiating discounts,” Baumer points out.

Some marketers claim that after paying the price of the name and the price for transmission, an e-mail campaign is on a par with the cost of a direct mail campaign.

“It may seem costly because I’m sending the message to much fewer people than I would with a direct mail piece,” observes Adriana Jones, marketing communications specialist at Made2Manage Systems Inc., an Indianapolis company that sells management software. “But my response is much higher, so it’s bringing down my cost per lead.”

That wouldn’t be true, of course, if the quality of lead is not very good. Unlike consumer e-mailadvertisements, which increasingly aim make a quick sale, B-to-B messages typically seek to build a long-term relationship.

Framingham, MA-based Staples.com builds loyalty by offering services such as e-mail reminders of when a customer needs to reorder office supplies or dollars-off coupons for spending a certain amount of money. “We’re always searching to build brand recognition and build upon the relationships with our customers,” says Staples.com spokesperson Debbie Hohler.

A typical way companies reel in prospects is by sending a series of messages. Laura Kuras, list director of Thomas Business Lists, which rents lists and markets its registries of manufacturers directly to business people, mentions a technique her list clients use. The first e-mail introduces the company. The second contains a newsletter about the recipient’s industry and indicates how the vendor’s product is beneficial to that industry. The third specifically addresses how the product can help the recipient’s industry and specific job, and may offer a premium if the recipient visits the Web site and shows a willingness to be contacted by phone.

Developing a good-quality lead is dependent on selections, such as type of industry, job area and number of employees at the company. Geographic selects, for example, are vital to a seminar company staging talks in cities around the country.

E-mail is a terrific relationship-building tool because it is fast. “You can be a one-man show and send an e-mail out and have business in 24 hours,” says Kuras. “If you do it with direct mail, you have to pay for costs and then sit there for six weeks wondering how it did.”

Job functions are a vital selection to have, Made2Manage discovered in tests this spring. “A test that went to CEOs received four times less response than a campaign that went to a mix of different job functions,” says Jones.

Made2Manage has also been testing different offers. A transmission offering a free T-shirt with a humorous Letterman-esque Top 10 Reasons to Purchase Made2Manage Software message delivered a clickthrough of 4% and a lead generation of almost 9%. But a gift book about leadership techniques delivered less than 1% lead generation.

Perhaps the T-shirt offer, “Visit our Web site today for free T-shirt,” did well because the offer was at the top of the message. Tests show that business people are too busy to slog through the message to get to the offer. “The entire message should be three to four paragraphs, or at least make sure paragraphs are choppy and small,” says Jones. “I want people to be able to get the gist of the offer.”

A clear, concise subject line is equally important, because that is what convinces the recipient to open the message. “Free T-Shirt” might work in the subject line, but that’s a risky strategy, say marketers, since companies are increasingly filtering out the word “free” from their servers.

The hotlink to the Web site should also be near the top. “If they are interested, they will react within the first couple of seconds,” observes Kuras. “I’ve seen messages where they put the hotlink at the end, and their response rate is abysmal.”

Finally, marketers are learning to limit the number of messages they send each recipient to between two and five a month. Given the scarcity of names, marketers don’t want create list fatigue. “The list pool is going to grow,” says Wexler, “but you’re already seeing people more reluctant to give their name out.”