A01 chaf6542 06 se fm indd

2 

Arbitrary divisions of customer database. This approach is also useful since the mar-

keter can set thresholds of value relevant to their understanding of their customers.

 

   For example, RFM analysis can be applied for targeting using email according to how a 

customer interacts with an e-commerce site. Values could be assigned to each customer as 

follows:

Recency:

1 – Over 12 months

2 – Within last 12 months

3 – Within last 6 months

4 – Within last 3 months

5 – Within last 1 month

Figure 9.22

RFM analysis

Frequency

5

Highest

Each R 

quintile contains

20% of all customers

R = 5, F = 5 contains

10% of customers

Lowest

Note here boundaries are arbitrary in order to place an equal number into each group

4

3

Direct mail

Phone

Email/web

only

2

1

Monetary

5

4

3

2

1

Recency

5

4

3

2

1

M09_CHAF6542_06_SE_C09.indd   452

7/23/14   1:29 PM

---

453

Chapter 9  Customer relationship management

Frequency:

1 – More than once every 6 months

2 – Every 6 months

3 – Every 3 months

4 – Every 2 months

5 – Monthly

Monetary value:

1 – Less than £10

2  –  £ 10–  £50

3  –  £ 50–  £100

4  –  £ 100–  £200

5 – More than £200

Simplified versions of this analysis can be created to make it more manageable, for 

example a theatre group uses these nine categories for its direct marketing:

Oncers (attended theatre once)

● 

Recent oncers 

attended <12 months

● 

Rusty oncers 

attended >12, <36 months

● 

Very rusty oncers 

attended 36+ months

Twicers:

● 

Recent twicer 

attended <12 months

● 

Rusty twicer 

attended >12, <36 months

● 

Very rusty twicer 

attended in 36+ months

2+ subscribers:

● 

Current subscribers 

booked 2+ events in current season

● 

Recent 

booked 2+ last season

● 

Very rusty 

booked 2+ more than a season ago

Product recommendations and propensity modelling

Propensity modelling

 is one name given to the approach of evaluating customer character-

istics and behaviour, in particular previous products or services purchased, and then making 

recommendations for the next suitable product. However, it is best known as ‘recommend-

ing the “ next-  best product” to existing customers’.

A related acquisition approach is to target potential customers with similar characteristics 

through renting direct mail or email lists or advertising online in similar locations.

The following recommendations are based on those in van Duyne et al. (2002):

1 

Create automatic product relationships (i.e.  next-  best product). A  low-  tech approach 

to this is, for each product, to group together products previously purchased together. 

Then for each product, rank product by number of times purchased together to find 

relationships.

2 

Cordon off and minimise the ‘real estate’ devoted to related products. An area of screen 

should be reserved for ‘ next-  best product prompts’ for up-selling and  cross-  selling. 

However, if these can be made part of the current product they may be more effective.

3 

Use familiar ‘trigger words’. This is familiar from using other sites such as Amazon. Such 

phrases include:

‘Related products’, ‘Your recommendations’, ‘Similar’, ‘Customers who bought . . .’  

‘Top 3 related products’.

4 

Editorialise about related products, i.e. within copy about a product.