"Dave Hauslaib" wrote
>The "recency" category seems relative. Should a customer
>get a better score if she just purchased from you than if
>she purchased a long time ago?
I'd say that RFM values need to be interpreted & used in the
context of the industry, market & product sector in which a
company operates
>if you are selling used cars, do you really want to
>target a customer for purchase that just bought from you?
>Chances are he is not going to need another car right away,
>but someone who bought from you three years ago might be in
>the market to buy another one.
Common-sense would indeed say that a customer who bought say
3 years ago is more valuable than some-one who bought last
month as they are more likely to be back in the market for a
car. This doesn't undermine 'recency' as a modelling &
selection value for a marketing communication, but reinforces
it as it's the 'long' length of time since last purchase that
is important & would contribute to a higher score & therefore
selection relative to other customers (or prospects). Also,
frequency of purchase is also likely to be important since
someone who has bought a car frequently over a period of
time is more likely to buy again
>If you are selling single-use contact lenses,
>then sure, a customer that just bought
>from you is more likely to repurchase the same item
In this example, frequency of purchase is likely to be more
important than recency as disposable contact lenses are
generally low price frequent purchases compared to cars -
recency is still important though - if only to tell you
they're still in the marketplace for lenses & haven't, say,
bought spectacle or gone elsewhere for lenses
Only way to know for certain the importance of each of the
RFM values for any company with a customer 'data set' that
can be used for RFM analysis is to actually do some
analysis & modelling. Once this has been done then a
strategy (or different strategies for different segments)
can be devised.
I've so far read 1/2 of Jim Novos's book 'Turning Customer
Data into Profits' - it's covers RFM in non-techie language
& is as straight forward a guide as I've come across - at
least for non-statistician marketers like myself.
Fortunato Licata
Book Clubs Associate
flicata_at_bca.co.uk
Received on Fri Mar 08 2002 - 19:55:13 CST
