Grouping customers into different RFM categories

 

Grouping customers into different RFM categories

In the examples above, each division for recency, frequency and monetary value is placed 

in an arbitrary position to place a roughly equal number of customers in each group. This 

approach is also useful since the marketer can set thresholds of value relevant to their 

understanding of their customers.

RFM analysis involves two techniques for grouping customers:

1 Statistical RFM analysis

This involves placing an equal number of customers in each RFM category using quin-

tiles of 20 per cent (10 deciles can also be used for larger databases) also shows one application of RFM with a view to using communications 

channels more effectively. Lower-cost e-communications can be used to correspond with 

customers who use only services more frequently since they prefer these channels, while 

more expensive offline communications can be used for customers who seem to prefer 

traditional channels.

2 Arbitrary divisions of customer database

This approach is also useful since the marketer 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

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 oncer attended 612 months

● Rusty oncer attended 712 but 636 months

● Very rusty oncer attended in 36+ months

Twicers:

● Recent twicer attended 612 months

● Rusty twicer attended 712 but 636 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 charac-

teristics and behaviour, in particular previous products or services purchased, and then 

making recommendations for the next suitable product. However, it is best known as 

recommending the ‘Next best product’ to existing customers.

A related acquisition approach is to target potential customers with similar character-

istics 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. (2003):

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’. That 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. That is, within copy about a product.

5 Allow quick purchase of related products.

6 Sell related product during checkout. And also on post-transaction pages, i.e. after one 

item has been added to the basket or purchased.

Applying virtual communities and social networks for CRM

We discussed some of the psychological reasons for the popularity of social networks in 

Chapter 2 in the section on consumer buyer behaviour and in Chapter 9 we will review 

some of the related Web 2.0 marketing techniques that can be used for customer acquisi-

tion. But in this section, we consider why social networks have developed and how they 

can be used to develop customer understanding and for relationship building.

The reasons for the popularity of virtual communities today such as the social networks 

Facebook, Google+ and LinkedIn can be traced back to the nineteenth century. The 

German sociologist Ferdinand Tonnies (1855–1936) made the distinction between public 

society and private community (Loomis, 1957). Tonnies employed the terms Gemeinschaft, 

meaning ‘community’ (informal, organic or instinctive ties typified by the family or neigh-

bourhood), and Gesellschaft, meaning ‘society’ (formal, impersonal, instrumental, goal-

orientated relations typified by big cities, the state and large organisations). Membership 

of Gemeinschaft is self-fulfilling (intrinsic motivation), whereas being a member of a 

Gesellschaft is a means to further individual goals (extrinsic motivation).

Marshall McLuhan (1964) posited that ‘cool’ (meaning on-going and shared) and inclu-

sive ‘electric media’ (meaning telephone and television, rather than books) would ‘retribalise’ 

human society into clusters of affiliation. Nicholas Negroponte (1995) predicted that in the 

near future ‘we will socialise in digital neighbourhoods’. Manuel Castells (1996) has devel-

oped the concept of ‘networked individualism’, in which individuals build their networks 

online and offline on the basis of values, interests and projects, and believes that ‘our socie-

ties are increasingly structured around the bipolar opposition of the Net and the Self.