A terrific article from CMO Innovation highlights several best practices for designing and implementing a successful retail loyalty strategy. Apparel manufacturer The North Face found themselves facing a particularly thorny challenge: How do you encourage repeat purchase when your products are too well made?
The entire CMO Innovation article is worth reading for its insights on designing and implementing a successful retail loyalty programme. For loyalty marketers in any industry, a few key points stand out:
Understand your brand and business model. While The North Face has traditionally marketed their apparel products indirectly through retailers, growing its direct-to-consumer business has become a key component of growth. In devising its loyalty strategy, the manufacturer unearthed a core truth about its business model: its apparel products, mostly outdoor and active wear, are so well made that a customer may only buy, say, a fleece jacket once every five years or more.
That purchase cycle makes the brand more akin to an automobile, furniture, or luxury retailer than to a traditional high-frequency specialty retailer. This reality led The North Face to design a programme built more around engagement between purchases designed to encourage cross-sell.
Predictive analytics are key to success. While the manufacturer performed segmentation based on traditional RFM metrics, they quickly learned that predicting customers' future outdoor and leisure activity was critical to success. Money quote:
"[The North Face] team learned that traditional analysis and segmentation tools were not necessarily effective for their purposes. 'We needed to understand customers' behaviors and set goals based on the type of outdoor activities our customers wanted to do,' [said Ian Dewar, director of CRM at The North Face.]
"Instead, the team started building campaigns based on customers' activities and product types relevant to those activities. Predictions on customers' future activities for a particular season had to be made to render the campaign more effective. To proceed with such predictions, a number of core item types were picked and divided into 30 categories, with a model created to perform correlation and predictive analysis based on customers' past purchases. Results were enlightening, to say the least."
Read the full story here.
- Rick Ferguson