Drilling Down Newsletter # 23 - August 2002 -
Latency Models, RFM, LifeCycles
Drilling Down - Turning Customer
Data into Profits with a Spreadsheet
Customer Valuation, Retention,
Get the Drilling Down Book!
Now also available online through
Amazon and Barnes & Noble
The Marketer's Common Sense Guide to E-Metrics - 22 benchmarks to understand the major trends, key
opportunities, and hidden hazards your web logs uncover.
To find out more about this topic:
In This Issue:
# Topics Overview
# Best of the Best Customer Marketing Links
# Latency: Services, Non-Profits, Distribution
# Questions: RFM and Customer LifeCycles
Hi again folks, Jim Novo here.
August is the month I try to keep the newsletter short. But
I'm not going to do what the trade press does and trot out a bunch of
lame content. Nope, this issue is covers various Latency models
and sets up our next topics - Recency and RFM models - with a question
from a confused fellow Driller.
A special note: I'm on a panel at the upcoming Search Engine
Strategies 2002 in San Jose, CA
Perfecting Paid Listings
Tuesday, August 13, 1:15pm - 2:45pm
This is an "advanced class," for people who have some
experience with paid listings but are looking to really crank up the
ROI on them. Copy, tracking methods, tips and tricks, all up for
grabs. Yours truly is of course addressing metrics and
tracking...more info here.
If you're around the show, please try to say Hi, I always like to
put a face with the typing!
OK - Let's do some Drillin'!
Best Customer Retention Articles
This section flags "must read" articles moving into the paid trade
magazine archives before the next newsletter is delivered. But
in July, the trade media realizes nobody is paying attention anyway
and pulls out all the "filler stories" from the files.
I figure you've had enough "360 degree view" crap to last a
lifetime at this point, so no links this month.
Tracking the Customer LifeCycle:
Real World Examples
Note: If you are new to our group and want to know more about the following ongoing discussion, the background theory is
The prior examples: Hair Salon and B2B
Latency: More Real World Examples in
Services, Non-Profits, and Distribution
Some businesses look on the surface like they many not be suitable
for using behavioral profiling. Take the electric power
business. The customer bill is just about the same every month,
with seasonal variations, and it's not like the customer has a choice
as to whether to buy power each month. Yet they can defect to an
alternative provider, and before you know it - it's too late. The cellular
biz has similar attributes,
particularly with the growth of annual contracts and bulk rate
minutes. How do you know when a telecommunications customer is about to defect?
Or take the insurance business. A very long cycle business
with very little transactional activity. You sign up for
insurance, get billed once or twice a year, and that's it. How
do you profile behavior in a business like this, where the customer
may be around for 5 years and then all of a sudden, just defects?
The answer is you profile activity other than the revenue related
activity, for example, calls to a phone center or visits to a web
site. When you look at "best" customers vs.
"worst," are there calling or visiting patterns which stand
The key is to group customers by "best" and
"worst" and look for any pattern that separates the two,
then use behavioral profiling methods to detect the likelihood of
defection. If you have a website or telephone "self service"
interface, falling use of it might mean customers are getting ready to
defect, or it might mean they are satisfied and are going to stay long
term. In situations like this, simply find "known
behavior" (actual defectors and best customers) and then
compare. This exercise should allow the creation of
actionable trip wires for the business.
In a long cycle, low transaction frequency business like insurance,
you may have to extend your time horizon to pick up enough meaningful
transactions. Instead of looking at behavior on an annual basis, you might have to look over a 3
year or 5 year period.
There's no way to tell in advance what these metrics will be, but
the customer behavior will "speak" and tell you which data
In the non-profit world of donors instead of customers and
donations instead of purchases, there is more emotional attachment
involved with the behavior. It pays to focus attention on
Latency by average donation amount, for example.
A Latency trip wire for a small, impulsive donor is likely to be
shorter than that of a large, emotionally attached donor. If you
try to repeat a large donor too soon after a substantial donation, you
risk looking a bit greedy. The only way to discover the
appropriate Latency trip wire for different types of donors is to
Set up some average donation "ranges," rank donors with
these ranges, and see if you can discover the best trip wire for each
range. You might also look at LifeTime donations in the same
way, creating a second ranking or "score." If
you rank donors this way, you can up with a two variable scoring
system that maximizes response and donation value over time. The
timing of your mailings would be triggered by some combination of
average and LifeTime donation scores. For example, a large
average donation donor with low LifeTime donations could be approached
more frequently than a large average donation donor with high LifeTime
Distribution or Supply Chain mechanics can benefit a great deal
from Latency trip wires, because you have built-in data points.
If you think about it, re-order or stock-out points are a type of trip
wire, but they are focused on the end user. What about the
supplier? The supplier should have their own set of trip wires,
tracking the average length of time between orders from end
users. When the order doesn't come though, something is going
on. Usually by the time a supplier realizes the order is late,
they have lost the contract.
This same idea applies to "soft distribution" methods
such as agent relationships. Let's say you originate insurance
or financial products that are sold by independent agents. These
agents also sell the products of other originators. Do you have
trip wires on these agents, so that an agent who is writing five whole
life policies a month that suddenly drops down to one a month is
identified and communicated with? Seems like it would be a good
idea to me, and not very difficult to do.
So that's it for the our Real World Examples using Latency.
Latency is a valuable tool but it can be somewhat of a fuddy-duddy;
generally a more reactive metric than a predictive
one. Next month we'll move on Recency
- the most powerful of the single-variable predictive metrics and the
doorway to RFM - the Swiss Army knife of simple behavioral models. -----------------------------------------
I can teach you and your staff the basics of high ROI
customer marketing using your business model and
customer data, and without using a lot of fancy software. Not ready for the expense and resource drain of CRM?
Get CRM benefits using existing resources by scheduling
Questions from Fellow Drillers
Jim's note: If you still don't know what RFM is
and how it can be used to drive increased profitability in almost any business, read
Q: I have a small sampling of the RFM scores that correspond to
the various lifecycle stages. For instance, 111 & 112
correspond to the acquisition stage, 333 & 443 to the growth
stage, etc. However, I'm looking for a complete listing of all
125 possible RFM scores and their corresponding lifecycle stages.
Can you please send this my way?
A: Wow, I certainly hope you didn't get this idea from me; if you did,
I have done a terrible job of explaining something somewhere. I
would be very interested in the source of this idea, that a LifeCycle
stage can correspond to a single RFM code or score.
An RFM code or score is the ranking of a single customer against
all other customers for likelihood to respond and future
value. High scores equal high future value; low scores equal low
A single RFM score represents this ranking at a fixed point in
time - the day the scores were created. There is no
"cycle," which implies "over time," inherent in an
RFM code. Only if you knew the previous code or sequence of
codes could you imply a "LifeCycle stage." This is, of
course, what my
book is about - using a modified version of RFM to track and
profitably act on customer LifeCycle behavior. If you know the
LifeCycle, you can predict behavior. If you can predict
behavior, you can dramatically improve marketing ROI.
If a customer is a 333, you don't know if they are falling or
growing into it. They could be coming from above it - falling in
value, or coming from below it - rising in value. For example,
most new customers start at a 51x - they have to, because by
definition, they are "new" (R = 5) but have bought once (F =
1). But this same customer 3 months from now might be a 555 or a
222 - either ramping up or sliding into oblivion. If you don't
know what their score used to be, you can't imply
anything about a "cycle" or any "stage" in the
relationship with the customer.
That said, customers in the 111 and 112 are typically old, defected
customers - not new or "acquisition stage" customers as you
said. All customers start in the high numbers and work
their way down into the low numbers throughout their lifecycle.
The question is how long will it take to get from high to low, and can
you do anything to slow this process or stop it. The scores tell
you if what you are doing is working, and how to drive profitability
following the two fundamental rules of High ROI Customer
1. Don't spend until you have to
2. When you spend, spend at the point of
If you are looking for some generalized system, I wouldn't worry
about the detail of 125 RFM codes, there is really no meaning there
unless you have millions of customers. The most important
variable, from a LifeCycle perspective, is usually Recency, so you
could roughly categorize the LifeCycle of customer into 5 blocks using
the R score. The second two variables, F and M, have nothing to
do with the lifecycle of the customer, but the value of the customer.
These are two different issues. Any customer with a low R value
but high "FM" value is a very valuable customer that isn't a
customer anymore. In terms of Lifecycle, they are at the
end. In terms of value, they are at the top.
If you are a consultant, agency, or software developer with clients needing action-oriented customer modeling or High ROI Customer Marketing program designs,
If you are in SEO and the client isn't converting the additional
visitors you generate, click here.
That's it for this month's edition of the Drilling Down Newsletter. If you like the newsletter, please forward it to a friend - why don't you do this now while you are thinking of it? Subscription instructions are at the top and bottom of the newsletter for their convenience when subscribing.
Any comments on the newsletter (it's too long, too short, topic suggestions, etc.) please send them
right along to me, along with any other questions on customer Valuation,
Retention, Loyalty, and Defection right here.
'Til next time, keep Drilling Down!
- Jim Novo
Copyright 2002, The Drilling Down Project by Jim Novo. All
rights reserved. You are free to use material from this
newsletter in whole or in part as long as you include complete
attribution, including live web site link and/or e-mail link. Please
tell me where & when the material will appear.