Business Resource Center
Subscribe

KIPLINGER RECOMMENDS

Home > Customer Service, Growing Your Business
 
 

EXECUTIVE POLL

How likely is it that we’ll experience a global depression like in the 1930s?

It’s a certainty
There's a good chance
Unlikely
No chance
Not sure
 
   view results
ADVERTISEMENT
 
 

OUR PREMIUM CONTENT


The Kiplinger Letter
 
 
 

CURRENT LETTER

 
The Kiplinger Washington Editors
Oct. 10, 2008
 

Stock Market Panic:
What Happens Next?

A heart-stopping, gut-wrenching stock market plunge is classic panic. It'll end eventually, but the economy will still need to work through a recession. This week's Kiplinger Letter looks at how we see the economy and government moves to shore up credit markets unfolding in the months ahead.
 
YOUR FEEDBACK
SUBSCRIBERLOG: Got a topic you'd like to discuss? Or a problem or question? Please join our exclusive forum for Letter subscribers only.
 
ASK US: A Kiplinger Letter editor will promptly answer subscriber questions.
 
 
OPEN FORUM: Share your insights and analysis with other visitors.
 
About a year ago I started a golf accessory online business . I would like to know how I can best market the site to get more visibility from customers as well as differentiating myself from other golf online store.
-- wyngategolf
 

A Simple Framework for Successful Loyalty Program Design

 
 
Mickey Neuberger
Loyalty Lab Inc.
Mickey Neuberger is senior director of loyalty strategy at Loyalty Lab Inc., a customer management and brand management company that specializes in designing and implementing customer loyalty programs. Its clients include Bally Total Fitness, 1-800-Flowers.Com, TicketsNow and Smith & Hawken.

As with any marketing discipline, designing a loyalty program is both a science and an art. The science concerns nailing the program economics and mechanics, such as how customers enroll, how members accrue points and how the benefits/rewards are structured. The art is the marketer's opportunity to differentiate the program, aligning it with its brand and customers.

For example, the basic program mechanics of Southwest Airlines Rapid Rewards are that eight round-trips will earn you one free round-trip. However, Rapid Rewards is much broader. Program nomenclature (e.g. "Freedom Awards"), special companion passes, surprise and delight rewards and communications, including free drink coupons and birthday cards, all align with Southwest's fun and consumer-focused brand.

In this article, I focus on the former, the science of designing a loyalty program for primarily transactional businesses (loyalty in subscription or advertising business models needs to be addressed quite differently). I will provide a high-level framework to help marketers design programs based on their businesses.

Implicit in this discussion is that high-level corporate objectives for the program are twofold. The first is to drive increased revenue via a combination of higher retention rates, greater purchase frequency and larger average order size. The second is to develop a two-way dialogue between the brand and customer via increased intelligence and relevant content and/or offers.

Prerequisite: Analyze Customer Behavior

Aside from the high-level corporate objectives, detailed program design must be deeply rooted in a thorough analysis of consumer spending patterns and behaviors. This analysis provides loyalty program designers with a sense of which customer behaviors they will be able to and will want to influence. For example, if 90% of customers only make one purchase, then the loyalty program's sole objective may be to drive the second purchase. More specifically, in order to guide program strategy, marketers should understand customer data along the following dimensions:

• By frequency: How often are customers making repeat purchases? In a 12-month period, what percentage of customers is buying only once? Twice? Three or more times? What percentage of revenue does each of these segments represents?

• By revenue: How concentrated is revenue in the top customer tier? What percentage of total revenue does the top 10% represent? The top 25%?

• By engagement: How do customers segment by nontransactional engagement activities, such as posting reviews or to blogs or subscribing to a company newsletter? Which of these activities are indicative of higher member value (in terms of increased revenue and/or decreased costs)?

Armed with analysis across these three dimensions, marketers have the critical data they need to design successful loyalty programs.

Step 1: Pick a Program Based on Purchase Frequency

Using the table below, marketers should match their customer purchase frequency behavior, ranging from low to high frequency, to the corresponding loyalty program objective and design. Overall, the greater the frequency of customer purchases within a year, the further out the program threshold should be set. This, in effect, minimizes the erosion of margin while allocating reward funds to drive customers toward higher total customer spend levels.

Step 2: Structure Program Based on Revenue Concentration

It is common for the top 10% of customers to account for as much as 50% of total revenue. For this reason, it's important to recognize, thank and encourage these "best" customers. Therefore, effective loyalty programs usually recognize and reward these customers with an elite tier with increased service benefits (e.g. free shipping, concierge service), special offers (e.g. member-only sales), exclusive rewards (e.g. free tickets to a special event) and higher reward accrual rates (e.g. double points). Successful programs also leverage their evangelist base for referrals, providing low cost new customer acquisitions.

Most programs set the threshold for tiers at specific member-spend inflection points where a significant percentage of revenue is clustered within a small percentage of customers. In a three-tier model, the top tier's baseline member spend threshold should be equivalent to the total annual spend of the top 5%-10% of customers. The next best tier should be equivalent to annual spend of the next best 5%-10%.

Step 3: Encourage Engagement (if Important)

In instances where customers interact with brands in non-transactional ways, such as by posting a product review on the e-commerce channel or clicking on an e-mail, program designers should strongly consider rewarding that behavior. This is because there is usually a very strong correlation between engagement activities and long-term customer value.

For example, a customer who refers friends, opts out of print communication, registers for the e-newsletter, purchases gifts for others and uses the private label credit card will be more valuable in the long term than another customer who does not perform these actions but spends the equivalent amount. Therefore, designers should measure customer value based on nontransactional interactions as well as on all transactional activities, and then create a funding model that rewards members a percentage of all value-add activities.

Conclusion

By using the presented framework, marketers utilize data segmentation analysis to make informed program design decisions. This ensures that loyalty program mechanics are set up to influence both incremental and valuable customer behaviors. With the scientific part of program design fulfilled, marketers can shift focus to the more artistic creative elements, designing programs that embrace their core brand attributes and pique customer interest.

This summary was drawn from a white paper by Loyalty Labs. To read more about loyalty
programs at the company's blog, click here.

READER COMMENTS

Post a comment
 | 
Read all comments (1)


POSTED BY: Harold K. Wilkes (February 11, 2008 12:16 PM)
Thanks for the article. Step 2 is exactly what I've been trying to get across to our marketing director for years. Being a casino, we run a most valuable player club. We do a lot of scatter-shot promotions and get nothing measurable in return. Marketing loves to spend money but hates to be accountable. Maybe your article can help me get something going. Thanks.

SAVE, SHARE & DISCUSS:    |   |   |   |   |   |   |   |    
ADD HEADLINES: