July 15, 2012

Dear Catalog CEOs: Why Loyalty Programs Work, And Why They Don't Work

Dear Catalog CEOs:


A carefully crafted loyalty program can increase sales by somewhere around 10%, among customers who are deemed loyal.


And yet, for most of us, loyalty programs don't work.


At Nordstrom, good customers had an 85% chance of buying again in the next year, purchasing maybe ten times if the customer repurchases.  That's 0.85*10 = 8.5 expected annual orders per loyal customer.  If a loyalty program increases volume by 10%, then you achieve an incremental 8.5 * 0.10 = 0.85 orders per customer.


At your garden-variety catalog brand, good customers have a 45% chance of buying again in the next year, purchasing maybe two times if the customer repurchases.  That's 0.45*2 = 0.9 expected annual orders per loyal customer.  If a loyalty program increases volume by 10%, then you achieve an incremental 0.9 * 0.10 = 0.09 orders per customer.


Can you see the difference between the two business models?


It doesn't matter what the cataloger does ... a loyalty program is going to make no difference whatsoever.


It make all the difference in the world what Nordstrom does ... a loyalty program greatly increases sales and generates profit.


Trade journalists, bloggers, and many loyalty experts miss this subtlety in customer behavior.


Loyalty programs work in a high-frequency environment.


Loyalty programs cannot force infrequent buyers into frequent buying behavior.