Improve Customer Retention by Targeting Segments in Motion

The ever popular and highly predictive “RFM” metric provides a static point in time view of your customers by their purchase total and number of months since last purchase to date.  In general, more recent customers with higher dollars and frequency are more responsive than those who have not purchased in a while and spent less.  Knowing this is great for routine, broad communications, but the key to customer retention and cultivation requires the incorporation of additional, more tailored messages for groups who are in motion.

The Annual Revenue Equation

Customer retention is a critical component in the annual revenue equation which is derived from the total number of anticipated purchases and average spend from three main groups:

Existing Active Buyers + Inactive Customer Re-activation + New Buyers

The proportion of annual revenue coming from first time buyers vs. existing customers can vary greatly by brand and reflect heavily on the sustainability of the business model.   In a 50/50 scenario, half of annual revenue is generated from prospects and the other half is determined by existing customer activity in the current year.  Some brands are in constant replacement mode with more than 60%-70% of their annual revenue generated by first time buyers.  This is to be expected in a start-up scenario but in general, this is a less profitable business model for an established company because it costs more to acquire a new customer than it does to cultivate an existing buyer.  Conversely, brands that generate 80%-90% of annual revenue from their existing customer base, may be missing an opportunity for new customer growth.  Use your marketing database to understand where your brand falls on the spectrum.

Finding Customers in Motion

Most marketing databases provide a current snapshot for a given customer.  You may maintain detailed transaction history, but all of the summarized selectable fields are likely aggregated based upon the current update’s view.  For example, a customer has only one RFM segment based upon their purchase history to date.  In order to find customers in motion, you will first need to be able to select by key pre-aggregated fields as they looked at various points in time.  What time period variances should be examined?  The answer is data point specific and a balance between “the sooner the better so we can re-act” vs. “it takes time to measure a change”.   A good place to start is with the comparison of “This Month vs. Last Month” and “This Year vs. Last Year” on key data points.  Personas are more conducive to measuring movement over a longer period of time.  For example, a customer was in the top cluster/persona last year but this year they have migrated downward to a lower status.   While finding those who became first time buyers in the current month is something you will want to respond to more quickly.   Determine what you are willing to respond to and build fields around that.

 Top Segments for Targeted Messaging

Find groups that represent a significant population where you can develop a clear strategy.  Below are 5  segments to get you started:

  1. First time buyers are those who were acquired since the last database update.  Depending upon how quickly you are acquiring customers, you may need to look at a monthly acquisition group, even if you update more frequently.  First time buyers should be acknowledged and introduced to the brand.  The goal is to convert new cutomers to 2x buyers quickly.  The majority of customer value is often observed within the first 6 months of the relationship.  Timing is everything. 
  2. Re-activated buyers are those who made a purchase in the current month after a period of greater than 1 year of dormancy.  These customers are re-engaged and the goal is to keep them shopping.  Welcome them back to the brand and incent them to buy again.
  3. VIPs are your best customers.  They are not defined by their change in motion, but instead by their consistent activity that warrants attention to maintain a given level.  The definition of your best customer is a business decision.  When in doubt, take your top spenders that represent the 80/20 rule – those 20% of customers who represent 80% of revenue.  Alternatively, select those who have purchased 4x+ in the last two years.
  4. Newly at Risk customers are those who have just turned the corner in the definition of active to inactive.  Most often, this group can be defined as customers with 13 months since last purchase.  It is important to target them sooner than later for re-activation because the likelihood that they will make a purchase decreases as the recency of their last purchase increases.  The goal for this group is to “save before you lose them” with more aggressive incentives to purchase again.
  5. Upward Risers are customers who were in lower customer segments last year but have migrated to a higher customer value.  These newly created Top Customers need to be acknowledged for their increased loyalty and dedication to the brand.

Measure the Impact

The impact of targeted strategies can be easily quantified.  In its simplest form, purchase activity in response to a single email or wave of emails can indicate success.  Longer term impact measures can include increases in the overall customer rebuy rate of various targeted segments.  It costs less to generate a purchase from an existing customer than it does to acquire a new one.  A small increase in customer retention can reduce the burden on new customer acquisition to carry the annual revenue growth goal.

Elisa Berger, Ph.D.

About Elisa Berger, Ph.D.

Elisa Berger, Ph.D is Principal and Executive Vice President, Database Marketing at Cross Country Computer (CCC). Elisa has been successfully helping database marketers achieve their ROI goals for nearly three decades. She earned her Ph.D. in Applied Research at Hofstra University.
This entry was posted in Analytics, Data, Data Management, Direct Marketing, Email, Marketing Strategy. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *