Drilling Down Newsletter # 17 -
February 2002 - RFM, LifeCycle
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
Drilling Down Newsletter # 17 - February 2002
If you would rather jump to certain topics, use the "In this Issue" links below.
In This Issue:
# Topics Overview
# Subscribers Weigh In, Want Changes!
# Best of the Best: Customer Marketing Links
# Tracking the Customer LifeCycle:Examples
# Questions from Fellow Drillers
Hi again folks, Jim Novo here. This month, we have subscribers
demanding format and content changes, and some "expiring
links" to "must read" articles. Also up is Part
2 of the Beauty Salon Example (Manage), and a Real World question from
a practitioner who wants to "take a gamble" with Frequency
scoring ranges - how far back in time do you go when counting the
Frequency of transactions?
Let's do some Drillin'!
Subscribers Weigh In, Want Changes!
The first issue of the year brought a lot of responses from you to
my request for changes to the newsletter format and content.
Here are the top 3 requests with my responses to them:
Request: Provide HTML version so newsletter can be read as text or online, with direct links to the different sections of the HTML newsletter in the text newsletter.
Response: You got it, as you can see from the "In This Issue" section at the top of the newsletter. Sections are linked to an HTML version of the newsletter.
Request: Clarify the content of newsletter versus the website; how do subscribers keep up with "what's new?"
Response: Virtually all new content goes in the newsletter before it goes on the web site, either as a full text version
or a direct link to the new article. If you read the newsletter
each month, you are exposed to all the new content available. I consider newsletter subscribers, who to a great
extent have purchased my book (Thanks!) my primary
customers, and they get all new content first. Then the content goes up on the web site.
Request: Provide more "how to" examples and less vague theory.
Response: Oh, you hurt me so! It's difficult to provide meaningful examples without providing
the backbone theory that helps a reader understand the examples.
That said, what I will do is scale back on the theory and provide more examples for each
concept. So instead of 70% theory; 30% examples, we flip it over and go with 30%
theory and 70% examples. In effect, this change means we will cover each topic more
deeply and completely, but not cover as many topics in a year.
This tracks with the general comments I have received over the past
year on the newsletter, so I think people will dig it.
As usual, comments on format and content are appreciated!
Best Customer Retention Articles
This section flags "must read" articles moving into the paid DM News archives before the next
newsletter is delivered. If you don't read these articles by the date listed, you will have to pay
$25 to DM News to read them from the archives. The URL's are too long for the
newsletter, so these links take you to a page with more info on what is in the article
and a direct link to the article.
Note to web
site visitors: These links may
have expired by the time you read
can get these "must read" links e-mailed to
every 2 weeks before they expire by subscribing to the newsletter.
Attitudinal Data: CRM’s Crystal Ball
Read by: Expires February 22, 2002
Crystal Ball is a little bold to describe the value of survey data but the essence of this article is correct; survey data should be combined with "proof of validity"
using behavioral data.
Small Biz Has Big Loyalty Potential
Read by: Expires February 28, 2001
An often overlooked segment in loyalty, many small to mid-sized businesses offer an opportunity to grab share
using a few perks.
Small to mid-sized businesses might also take a look at the Simple CRM
program - CRM benefits without CRM costs. Click
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
Part 1 of the "Beauty Salon Example" is here.
A Tale of Two Hair Salons Part 2: Manage
Recall the events of Part 1 (Measure) - the owner of Salon B used a simple spreadsheet to look at the last visit date of best customers, and
found these issues with them:
- A substantial number of high value customers have not had an appointment in 6 months, about 20% of them.
- The average number of days between appointments is similar across
all the high value customers. It is, however, not the 30 days the
owner expected. It is more like 40 days.
The owner of Salon B has Measured the Customer Retention situation, and thinks:
I must be crazy for not looking at this before.
I would make more money by not cutting hair for a couple of hours a week if I used that time to get
even one of these high value customers to start making appointments
again. Now that I have
Measured this effect and know how
money it is costing me to not address the tardy Angela
need to Manage this situation somehow.
Over at Salon A, the owner knows the names of best customers who
"have not been in for a while." But this owner has no
system, no way to measure what the dynamics of the situation
are. How long is "a while"? But at Salon B, the
owner knows the average time between best customer visits is 40 days,
and there are customers in this group who have not had an appointment
in over 6 months. How can the owner get this business
back? The owner:
I'll just mail all these best customers who
have not had an appointment in over 6 months a postcard offering them
a discount. The postcards will say, "Since you are a best
customer, you are entitled to a 15% discount if you come in for a
visit within the next two weeks." They will come in and I
will start a new relationship with them, and find out why they have
not been in.
The owner of Salon B prepares the targeted postcards, mails them
out, and awaits appointments from these best customers.
The appointments never come.
A bunch of the postcards come back as "undeliverable,"
and the owner gets several phone calls from customers saying "I
now go to Salon A, take me off your mailing list."
Undaunted, the owner of Salon B reasons:
Clearly there is something wrong with this
approach. Best customers who have not had an appointment for 6
months must already be "defected" customers. They
obviously do not want to come back to me, and feel the relationship is
broken already. They have moved on and established new
I will try a new approach with the postcards,
and will use the same offer. But this time, I will mail the
postcards out as soon as the best customer has not been in for over 40
days. Since the average best customer comes in every 40 days, a
best customer who fails to do so is not acting like a best
So each week I will use my spreadsheet to
identify best customers who have not been in for 40 days, mail the
discount postcard out to them, and track the results.
After a month of mailing the weekly 40 day postcards to best
customers, the owner of Salon B sat down to analyze the program.
Of all the best customers mailed to, 25% had made new appointments,
and 75% had not. So in the short term, the owner had cut the 20%
best customer defection rate to 15%, because 1/4 of the best customers
called to make appointments at $150 each - minus the discount.
But even with the discount, the additional profits from these
customers paid for the postcard mailing many times over.
Despite this success, two things bothered the owner of Salon
B. The first was what customers who responded said when making
their discounted appointments. The second was the 75% of best
customers who did not respond. The owner thinks:
Half the customers who responded said to me,
"I'm so glad you mailed me a discount, I was planning on making
an appointment in the next week and would have made one anyway, so it
was great to get the discount." So I gave up margin and
profits I did not need to give up.
And how is it possible that so many of my
best customers never responded to my offer?
I wonder if there is a way to address these
two issues? If I could reduce the number of "would have
come in anyway" customers who got a discount, and increase the
overall response rate, I would be really making a ton of money on my
best customer retention postcard program. I have Measured
my best customer defection, and am Managing it with this
program. I wonder if there is a way to Maximize, to make
it even more profitable?
Next month, we'll check back in with the owners of Salon A and Salon B
and see what happens. Will Mary Lou ever show up at Salon
A? Is there a way to make more money in this program? Only
the data knows for sure...
Go to part 3 of the Beauty
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
Q: I am reading your eBook (up to page 67) and I am trying to
relate the material to work.
A: That's a good idea! Especially if work paid for the
book. If you paid for it, then work is lucky to have you....
Q: I am involved in Data Warehousing and I am looking at defining
some datamarts for our Sales people.
A: Perfect application for the book, will give you a
"common ground" to work from.
Q: Could you please help me with a question that has been
nagging me...Over what period of time should we calculate Frequency?
a) 30 days prior to the last sale
b) 12 months prior to the last sale
c) 36 months prior to today
d) All of the above
The data calculated by each method is different however they all look
A: If I can imply from your e-mail address that you work for
(large casino and sports betting organization), and the business we
are talking about is the casino / sports bet business....
Well, they all are useful, in their own way.
Generally, you want to use long cycle models with long cycle events
and short cycle models with short cycle events. For example,
retail purchases are pretty short cycle events, with supermarket
probably the shortest of them all. Enterprise Software and heavy
duty truck purchases are long cycle events. But within both of
these long cycle events, there are short cycle events, such as
software upgrades, replacing tires and exhaust systems, etc.
One way to "pin" the cycles down is to look at
inter-event Latency - the average number of days or weeks between the
events. If you are profiling casino visits and they happen on
average twice a year, the "cycle" or Latency is 6
months. Running an RF model at 30 days doesn't get you much
traction with a 6 month cycle. I would run it over 12 months, or
twice the cycle rate.
The closer you get to the actual latency in the cycle, the more
accurate the model can become, but the more erratic it can get over
disparate groups. This begs a question: can you split the
population and RF model each segment by itself? Sure!
Say you have a top 20% group, and these people visit on average
every 2 months. Then you have everybody else, who visit on
average 1x a year. So the model for best customer would be over
4 months (twice the cycle time) and the model for the others over 2
years (twice the cycle time). That way you adjust the model to
hang more closely with the behavior, and increase accuracy.
Q: P.S. Like your book.
A: Great, that's good to know. A little surprising, but
IT people like the book quite a bit. From what I gather, the
book helps IT people "get" database marketing enough to work
on requirements, and provides a tangible platform to discuss behavior
modeling issues with marketing folks.
Hope I answered your question, and feel free to keep asking -
you're a customer now!
That's it for this month's edition of the Drilling Down newsletter.
If you like the newsletter, please forward it to a friend!
Subscription instructions are top and bottom of this page.
If you are a consultant, agency, or software developer with clients
needing action-oriented customer modeling or High ROI Customer Marketing
program designs, click
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
'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.