HRE stands for Heterogeneity, Recency, Exit. The method is inspired from
stochastic modelling research in marketing from works of Morrison, Chen,
Karpis and Britney (1982). It uses individual purchase history data:
- to determine several customer profiles
- to infer customer's probability to become key given their profiles;
- to qualify the heterogeneity of this probability's distribution
among individual customers
- to forecast future customer profiles.
Recency, Frequency and Monetary are largely used criteria in direct marketing
and catalogue sales. This industry is rather prototypical for relationship
marketing and was probably historically the first to develop a direct customer
relationship.
We have suggested several RFM segmentation methodologies from simple
ones uniquely based on purchase history data to more complex ones including
explanatory variables and integrating Survival Analysis.
We have also operationalised and adapted a recent model
of direct marketing (Bitran & Mondschein, 1996) to reflect the
main mechanisms of relationship marketing communications.
Recency
Frequency
Monetary
Segments
in the
Home List
R4FM
R3FM
R2FM
R1FM
Mechanics of Direct Marketing
Home
List Mailing Lists
Rental Lists
The ideea behind the direct marketing model that can be generalised to
relationship marketing communications is that a well qualified customer
database (home list) is an asset that can be profitably managed. Creating
a home list is a costly investment, banks must buy (from the data base
market) potential customer lists (rental lists) that usually have low order
response rates, meaning that mailing and communication costs are higher
than returns from orders received. But respondents from rental lists are
included in the house list which has much higher response rates. These
are stimulated by specific direct marketing techniques. When the house
list becomes big enough it returns enough gains to compensate the losses
generated by the aquisition of rental lists. Aquiring rental list, although
costly, remains necessary in order to renew the house list by replacing
low return RFM customer segments with new entrants.
The distinction between "hard core loyals" and "potential switchors" helps
distinguish the customers who respond mainly to "defensive marketing" from
those who respond to "offensive marketing". This segmentation was first
used in marketing by Alfred Kuehn (1961) and more recently by Colombo and
Morrison (1989) and Bultez (1996 and 1997). The last author gave easy operational
solutions to both estimate and separate among customers the "hard core
loyals" from "switchors" and to evaluate econometrically, the effects of
"offensive marketing" on attractiveness. We have adapted and implemented
these methods in a computer program.